THE INFLUENCE OF WORKPLACE EMOTION-FOCUSED COPING NORMS By Catherine Ott-Holland A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Psychology – Master of Arts 2013 ABSTRACT THE INFLUENCE OF WORKPLACE EMOTION-FOCUSED COPING NORMS By Catherine Ott-Holland This study explored the extent to which workplace emotion-focused coping (EFC) norms exist in workplaces and whether these norms influence EFC behaviors and distal employee outcomes. EFC norms were expected to exhibit a proximal influence on the extent to which individuals use EFC behaviors to deal with negative emotions in the workplace. EFC norms were predicted to influence distal outcomes as well, such as workgroup satisfaction, physical strain symptoms, person-environment workgroup fit, and workgroup helping intentions. To assess this, 149 working adults from a number of different organizations filled out two online questionnaires with a week time lag between administrations. In the initial survey, participants rated the extent to which they perceive workgroup EFC norms. In the follow-up survey, participants filled out measures of both distal outcomes and the extent to which they typically engage in different EFC behaviors during negative events at work. Analyses indicate EFC norms are related to typical EFC behavior, but EFC norms and behaviors were not related to workplace outcomes. Copyright by CATHERINE OTT-HOLLAND 2013 DEDICATION Dedicated to my loving and supportive parents. iv TABLE OF CONTENTS LIST OF TABLES………………………………………………..……....…………….viii LIST OF FIGURES………………………………………………….…..…………..…....x INTRODUCTION………………………………………………………………………...1 Negative Affect...…………………………………………………………4 Emotion-focused Coping………………………………………………….5 Workgroup Norms for EFC: The Social Context of Coping……………...9 A Proposed Model of Workgroup Norms for Emotion-Focused Coping..12 Proximal Outcome: Affective Event Theory and Typical Coping Behaviors………………………………………………………...………12 Distal Outcomes………………………………………………………….14 Behavioral Mediation…………………………………………………….15 Self-Esteem as a Moderator……………………………………………...15 Trait impulsivity as a Moderator…………………………………………16 Gender as a Moderator…………………………………………………...17 Deep Acting and Surface Acting………………………………………...17 METHOD………………………………………………………….……….…….….…..20 Pilot study………………………………………………………………..20 Primary Study Participants and Procedures……………………………...21 Primary Study Measures…………………………………………………25 Primary Study Survey 1.................................................................25 Self Esteem........................................................................25 Trait Impulsivity................................................................25 Workgroup Emotion Focused Coping Norms...................25 Social support seeking norms (instrumental and emotionfocused)..............................................................................26 Coping through humor norms............................................26 Emotional expression norms..............................................26 Surface and deep acting norms..........................................27 Display Rules.....................................................................28 Demographics....................................................................29 Primary Study Survey 2.................................................................29 Typical EFC Behavior.......................................................29 Physical Strain...................................................................29 Workgroup Helping Intentions..........................................29 Workgroup Satisfaction.....................................................30 Workgroup Person-Environment Fit..................................30 RESULTS………………………………………………………..……..………….….…31 Sample Descriptives……………………………………………………...31 Factor Analysis of Descriptive and Injunctive Norm Items......................31 v Factor Analyses of Typical EFC Behaviors……………………………...33 Revision of Humor and Social Support Scales…………………………..34 Descriptive Statistics and Intercorrelations of Separate Situations...........34 Descriptive Statistics and Intercorrelations of Collapsed Situations.........38 Control Variables.......................................................................................39 Hypothesis 1...............................................................................................39 Hypothesis 2...............................................................................................40 Hypothesis 3...............................................................................................41 Hypothesis 4...............................................................................................41 Hypothesis 5...............................................................................................42 Hypothesis 6...............................................................................................43 Exploratory Analysis: Proximity as a Moderator of Norm-Typical Behavior Relations.....................................................................................43 DISCUSSION………………………………………………………………………..…..43 Emotions in Non-service Settings..............................................................44 Distinguishing EFC, Surface and Deep Acting, and Display Rules..........45 Situational Differences and the Distinction between Social Support Seeking-Emotional and Social Support Seeking-Instrumental..................47 The Relationship between Use of Humor Coping and Emotional Expression..................................................................................................49 Deep Acting: Unexpected Linkages..........................................................49 EFC Norms and Typical EFC Behavior Connections...............................50 Emotion-focused Coping Behaviors and Distal Employee Outcomes......53 Trait Impulsivity and Self-Esteem as Moderators of Norm-Behavior Relationships..............................................................................................54 Gender........................................................................................................55 Limitations and Future Directions.............................................................56 Concluding Remarks..................................................................................58 APPENDICES…………………………………………………………………….……..60 Appendix A: Pilot study consent form……………………………….…..61 Appendix B: Pilot study ratings…………….…………………………....62 Appendix C: Primary study survey 1 online consent form--management course version………………….………………………………..….........64 Appendix D: Primary study survey 1 online consent form--Mechanical Turk version....…………………………………………………..…….....65 Appendix E: Primary study survey 2 consent form-- management course version………………………………………………....…………………66 Appendix F: Primary study survey 2 consent form-- Mechanical Turk version………………………………………………………………........67 Appendix G: Self Esteem………………………………………………...68 Appendix H: Impulsivity………………………………………………...69 Appendix I: Workgroup Coping Norms…………………………........…70 Appendix J: Display Rules (Requirement to Hide Negative Emotions)……………………………………………………..……….....73 vi Appendix K: Demographics…………………………………….……….73 Appendix L: Coping Behaviors…………………………………….........74 Appendix M: Physical Strain Symptoms………………………………...76 Appendix N: Workgroup Helping Intentions…………………………....77 Appendix O: Satisfaction with Coworkers………………….…………...78 Appendix P: Workgroup P-E Fit………………………………..………..79 Appendix Q: Tables and Figures……………………………….....……..80 REFERENCES…………………………………………………..………….……....….118 vii LIST OF TABLES Table 1: Definitions of Coping………………………………………………......………81 Table 2: Emotion-focused Coping Dimensions…………………………………….........82 Table 3: Ratings of the emotional reactions to negative workplace situations…......……83 Table 4: Descriptive Information for Demographic Variables………………….....…….85 Table 5: Varimax Rotated Maximum Likelihood Factor Structure for All 25 Norm Items………………………………………………………………………………......….86 Table 6: Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 1…………………………………………………………………………………..……....87 Table 7: Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 2………………………………………………………………………………………......88 Table 8: Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 3…………………………………………………………………………………......……89 Table 9: Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 4………………………………………………………………………………………......90 Table 10: Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 5……………………………………………………………………………......91 Table 11: Descriptive Statistics of Manipulation Checks by Situation……………….....92 Table 12: Situation-Specific Typical EFC, SA, and DA Behavior Descriptives, Reliabilities, and Norm Intercorrelations…………………………………………….......93 Table 13: Situation-Specific Typical EFC, SA, and DA Behavior Inter-correlations with Outcome and Remaining Variables………………………………………………….......96 Table 14: Descriptive Statistics, Reliabilities, and Intercorrelations for Measures….......99 Table 15: Emotion-Focused Coping Behaviors Regressed on Workgroup EmotionFocused Coping Norms………………………………………………………………....103 Table 16: Emotion-Focused Coping Behaviors Regressed on Display Rules……….....104 Table 17: Outcomes Regressed on Typical EFC Behaviors…………………………....105 Table 18: Outcomes Regressed on Typical Surface Acting and Deep Acting Behaviors…………………………………………………………………………….....106 Table 19: Outcomes Regressed on EFC Norms……………………………………......107 viii Table 20: Outcomes Regressed on Surface Acting Norms, Deep Acting Norms, and Display Rules……………………………………………………………………….......108 Table 21: Coping Behaviors Regressed on Norms with Self Esteem as a moderator.....109 Table 22: Surface and Deep Acting Behaviors Regressed on Surface Acting, Deep Acting, and Display Rules with Self Esteem as Moderator………………………….....110 Table 23: Coping Behaviors Regressed on Norms with Trait Impulsivity as a moderator…………………………………………………………………………….....111 Table 24: Surface and Deep Acting Behaviors Regressed on Surface Acting, Deep Acting, and Display Rules with Trait Impulsivity as Moderator…………………….....112 Table 25: Outcomes Regressed on Emotional Expression with Gender as Moderator…………………………………………………………………………….....113 Table 26: Hypotheses and whether the data supported each…………………....……...114 Table 27: Coping Behaviors Regressed on EFC Norms with Physical Proximity as a Moderator……………………………………………………………………….....……115 Table 28: Surface and Deep Acting Behaviors Regressed on Surface Acting, Deep Acting, and Display Rules with Physical Proximity as Moderator………………….....116 ix LIST OF FIGURES Figure 1: Model of the Influence of Workgroup Norms for Emotion Focused Coping on Employee Outcomes…………………………………………………………………....115 Figure 2: Moderating effect of trait impulsivity on the relationship between norms for emotion-focused social support seeking and emotion-focused social support seeking…………………………………………………………………………………122 x INTRODUCTION Negative emotions are a fundamental component of everyday workplace experiences (Ashforth & Humphrey, 1995; Beal, Weiss, Barros, & MacDermid, 2005). How employees cope with these feelings plays a large role in their behavior at work (Rafaeli & Sutton, 1987; Fiebig & Kramer, 1998). Employees can choose to manage negative emotions in different ways. For example, imagine an employee learns he or she will not receive a long hoped for promotion at the end of the year. This situation is likely to elicit a negative emotion from the employee. That employee could manage this negative emotion in a number of ways. The employee could seek out colleagues so that he or she might vent about negative emotions, ask for advice, or obtain empathy or encouragement. The employee might try to find humor in the situation, and make jokes to feel better. The employee could express his or her anger and frustration by cancelling meetings or writing harshly worded emails to colleagues. Over time, general tendencies towards any of these coping behaviors could potentially have long term effects on individuals’ stress levels and work-related attitudes. Both the way employees cope with negative emotions at work and the norms surrounding emotion-focused coping (EFC) have organizational and societal relevance. For example, metaanalytic evidence has shown through seeking out social support, somewhat counterintuitively, has also show a positive association with negative health outcomes (r= -.04, CI lower= -.07, CI higher= -.004; Penley, Tomaka, & Wiebe, 2002). The different ways people cope with negative emotions have been linked to psychological and physical well-being (Affleck & Tennen, 1996), which have in turn been linked to organizational outcomes, such as the cost of insurance and the cost of illnesses that can be remunerated under worker’s compensation (Danna & Griffin, 1999). Given the fiscal concerns of organizations and the broader societal aim of worker well-being, 1 research studying how workers cope with their negative emotions at work has substantial practical value. The way that individuals experience and process negative emotions has been linked previously to the social context. For example, team research on venting has shown venting to members inside a team can produce agitation (Kowalski, 1996), whereas venting to non-team members can aid in problem sense-making (Volkema, Farquhar, & Bergman, 1996). Experimental research has shown confederates who outwardly express specific moods can influence the moods of participants in similar directions, exhibiting a process of affective sharing called emotional contagion (Barsade, 2002). Another element of the social context that may influence how individuals experience and cope with negative emotions is workgroup norms. Over time, norms tend to develop within workgroups (Tuckman, 1965; 1977; Kozlowski, Gully, Nason, & Smith, 1999). Norms influence members in a variety of ways. Norms influence how individuals perform tasks, perceive information, and interact with others (Kozlwski, Gully, Nason, & Smith, 1999). Over time, individuals develop shared perceptions in the workplace. These shared perceptions have a powerful influence over individual behavior (cf. Kozlowski & Bell, 2003). Given past research on how the social context influences individual affect, cognition, and behavior (cf. Banaji & Prentice, 1994), there is reason to believe that norms for EFC exist. Further, these norms may predict the extent to which individuals use different EFC strategies when negative events occur. If an employee observes others generally seek social support when experiencing negative emotions at work, that employee may be more likely to seek out social support following his or her own experiences of negative emotions. If an employee observes a norm of expressing feelings when colleagues are experiencing negative emotions, that employee 2 may be more likely to exhibit an emotional expression coping style at work when experiencing a negative emotion. Similarly, if an employee consistently observes others making jokes about feeling bad, that may influence the employee’s likelihood to cope with negative emotions using humor as well. The topic of norms for coping with negative emotions in the workplace fills an established research need in the organizational sciences. Barsade and Gibson (2007) pointed out the need for further research on the “affective culture” that exists within organizations. An organization’s affective culture captures processes beyond the well-established construct of emotional labor, defined as the display of expected emotions of service professionals in service encounters (Ashforth & Humphrey, 1993). Affective culture describes different types of norms surrounding emotions in a broader workplace context including, but not limited to, service duties. Among research on collective affect, the authors identified the normative influence of affective culture to be the topic least studied and most open to future development (Barsade & Gibson, 2007, p. 50). The purpose of this study was to determine whether EFC norms link to proximal and distal employee outcomes. Proximally, the framework of cognitive appraisal theories was used to assess how employees typically cope with negative affective events in the workplace. As will be described further in the following sections, it was expected that individuals would be more likely to engage in EFC behaviors that they view as normative in the workplace. Several distal outcomes of theoretical and practical significance, including workgroup satisfaction, physical strain symptoms, person-environment workgroup fit, and workgroup helping intentions were also examined. To the extent that EFC behaviors facilitate functional or dysfunctional outcomes, it 3 was expected that EFC norms would also predict positive or negative distal outcomes for employees. The following section briefly reviews existing research on affect, coping, and workplace norms. A novel construct, emotion-focused coping norms, is defined and distinguished from related norms. Research on the influence of normative perceptions is cited to explain how norms influence individual behaviors. Cognitive appraisal theory and Affective Events Theory (AET) are used to frame the proximal coping behaviors measured as part of the proposed study. Distal outcomes (workgroup satisfaction, physical strain symptoms, person-environment workgroup fit, and workgroup helping intentions) are further described and relevant hypotheses are made. Negative Affect Several frameworks for conceptualizing affect have gathered empirical support and are used widely within applied contexts (Watson & Clark, 1984, Ekman, 1992, Shaver, Schwartz, Kirson, & O’Connor, 1987). One of the most commonly used frameworks divides affective experiences into positive and negative dimensions (Watson & Clark, 1984). Positive affective experiences are described as pleasant, energetic feelings such as happiness and joy. Negative affect represents unpleasant feelings such as anger, anxiety, guilt, and sorrow. Whether or not positive and negative affect exist on a singular continuum has been debated, although evidence seems to favor the existence of two largely orthogonal dimensions (Tellegen, Watson, & Clark, 1999; Diener & Emmons, 1984). In addition, affect is differentiated into two types of experiences: moods and emotions. The most distinguishing aspect of emotions is that they require an appraisal of an event. Individuals generally make causal attributions when experiencing an emotion (e.g. being angry at someone). Moods, in contrast, lack obvious antecedents and are generally more diffuse and 4 less intense (Weiss & Cropanzano, 1996). Whereas emotions can strongly determine how an individual responds to a situation, moods evoke a broad array of behavioral responses that are less dependent on whatever prompted the experience (Weiss & Cropanzano, 1996). The present study focuses specifically on work related events that elicit anger, sadness, fear, distress, and guilt. These fundamental negative emotions are represented in both Watson & Tellegen’s (1985) PANA model and Eckman’s (1999) framework of emotions. To understand how these emotional experiences are processed and translated into behavioral responses, the present study draws upon the EFC literature, and specifically, cognitive appraisal theories of coping. Emotion-focused Coping The coping literature within the field of psychology is expansive and lacks a clear consensus on how to categorize dimensions of coping (see Skinner, Edge, Altman, & Sherwood, 2003). Problem-focused coping (PFC) and emotion-focused coping (EFC) provide a commonly used distinction in the coping literature. These two types of coping were originally put forth by Folkman and Lazarus (1980) and further elaborated in Carver, Scheier, and Weintraub’s (1989) development of the COPE measure. Whereas PFC describes how individuals try to change or eliminate the source of stress, EFC focuses on what individuals do to minimize the emotions associated with a negative event. In the present study, EFC was of particular interest. Carver et al. (1989) describe EFC as “…aimed at reducing or managing the emotional distress that is associated with (or cued by) the situation.” This definition was deemed insufficient for the present study for two reasons. First, this definition does not make clear whether it refers to behavioral or cognitive coping efforts. Given this study’s emphasis on social norms, behavioral manifestations of EFC must be emphasized as they are likely to inform perceptions of norms. Although individuals do cope with 5 negative emotions in cognitive ways (e.g., reappraising the situation), the information people obtain about how others cognitively process emotions is likely to be indirect or inferred. For example, unless George tells his colleagues that he “thinks about the positive side of things” when workplace problems arise, their estimation of whether this is true is mostly inference. However, if George cries at his desk when he cannot get his computer code to run, his colleagues now have some information about how George copes behaviorally with negative emotions in the workplace. Thus, behavioral information are critical for workgroup members to perceive there is a social norm, and this should be reflected in the present study’s definition of EFC. Another reason why Carver et al.’s (1989) EFC description needed further clarification was that it did not specify which negative emotions are of interest. For example, does “distress” refer to guilt? Or anger? To generate an improved operational definition of EFC for this study, a review of global definitions of coping were gathered from the existing literature. The findings are featured in Table 1. The definitions in this table show clear conceptual overlap. All describe coping as a reaction to an upsetting event. All emphasize some form of effort is made on the part of the individual. The present study adopted a definition of coping based on Folkman and Lazarus’s (1980) definition. Coping was defined in this study as behavioral efforts to master, tolerate, or reduce external and internal demands and conflicts among them. The definition fits the purposes of the proposed study, as it emphasizes behavioral efforts that more readily translate into norms than cognitive efforts. Furthermore, the language does not restrict coping to the realm of stress, but instead uses “external and internal demands” to describe the phenomena that elicit both strain and negative emotions. Much of our understanding of EFC stems from cognitive appraisal theories (Carver et al., 1989). Cognitive appraisal theories provide a basis for understanding how individuals analyze 6 emotion-related information and decide upon coping strategies (Carver et al., 1989). Lazarus’s (1966) cognitive appraisal model divides reactions to stressful situations into primary and secondary appraisals and coping behaviors. The primary appraisal involves the basic decisions as to whether a situation is relevant to the individual’s goals, and if so, whether or not it poses a threat. If in the primary appraisal the individual identifies the situation to be threatening, that individual must make a secondary appraisal of what that threat means given the situation and how to react behaviorally to the stressor. Recall the initial example of an employee who does not receive an expected promotion. That employee’s primary appraisal was that the organization’s actions were in some way threatening or stressful. The employee may feel disenfranchised by the decision, or may worry that his or her hard work will go unrecognized by the organization. This appraisal is the beginning of the employee’s experience of a negative emotion, such as fear, anger, or sadness. The secondary appraisal is the decision the individual makes as to how they will react and potentially manage this negative emotion. Following the secondary appraisal, the individual engages in the coping behavior selected as most viable in the secondary appraisal. In this study, EFC norms focus on the collective effect of the types of emotion-focused coping behaviors generally exhibited in a workgroup, and proximally, whether those norms have an influence on the EFC behaviors of individual group members. Although these norms will be further described in the section that follows, coping behaviors (and not primary or secondary appraisals) are emphasized in this study because their behavioral nature makes them visible and distinguishable to others. In Carver, Scheier, and Weintraub’s (1989) framework, not all coping categories are coping behaviors; many are part of the primary or secondary appraisal. For instance, the positive reinterpretation and growth 7 dimension, defined as redesignating a stressful situation in positive terms, is considered a form of primary appraisal. The planning dimension, defined as thinking through how to cope with a stressor, is said to occur in the secondary appraisal process. Primary and secondary appraisals are cognitive processes, and thus, were not included in this study. Instead, the focus was on coping behaviors, such as seeking social support. Three types of EFC strategies were selected for the present study’s hypotheses. These definitions are listed in Table 2. The first type of behavioral EFC coping included two subdimensions: seeking social support for instrumental reasons and seeking social support for emotional reasons. Although seeking social support for instrumental reasons is typically considered a PFC strategy, these two subdimensions were at first grouped in the present study given individuals may simultaneously seek out social support for both emotional and instrumental reasons. Furthermore, these subdimensions are highly related in past research (r=.69, Carver et al., 1989). The second dimension of behavioral EFC was the use of humor, representing the extent to which individuals react to stressful situations by making jokes. Past research has shown humor may be used intentionally to help cope with stressful events (Martin & Leftcourt, 1983; Stanton, Kirk, Cameron, & Danoff-Burg, 2000). As the use of humor can be behavioral in nature, there is reason to believe this may form a norm. The third dimension of behavioral EFC was emotional expression. Although this dimension is defined as both interindividual and intraindividual, the examples of intraindividual coping in the definition (i.e., journal writing, artistic production) are behaviors. Thus, expression can be viewed as a behavioral dimension of EFC. 8 The present study predicts norms for EFC using these types of behaviors will have a direct effect on individual coping behaviors. In the next section, workgroup EFC norms will be defined and further described. Workgroup Norms for EFC: The Social Context of Coping Research has shown individuals’ thoughts and actions are both heavily influenced by the social context (e.g. Banaji & Prentice, 1994). In the workplace, norms for how employees generally deal with negative emotions may influence how individuals cope with negative emotions. Cialdini, Reno, and Kallgren (1990) argue that norms are often used to describe two semantically different concepts: descriptive norms and injunctive norms. They describe descriptive norms as what occurs in general, or on average. Injunctive norms constitute the rules or expectations surrounding what is and is not appropriate in a social context. The present study is interested in the effects of descriptive norms for employee EFC. This emphasis on descriptive norms surrounding emotion regulation at work extends prior research which, as will be discussed, has generally focused on injunctive norms. The construct most similar to this idea of workgroup EFC norms within the field of industrial and organizational psychology is display rules. Display rules are defined as norms about appropriate emotional expression in a given situation (Schaubroeck & Jones, 2000). The definition’s focus on the “appropriateness” of emotional expression identifies this as an injunctive norm. The literature surrounding emotional display rules is often studied in the context of emotional labor, defined as the display of expected emotions of service professionals in service encounters (Ashforth & Humphrey, 1993). Display rules tend to emphasize employee obligations to express a certain affect in specific situations at work (Diefendorff & Richard, 2003), and fail to capture the workplace norms for dealing with negative affect that occur in 9 situations where prescribed emotion-related expectations are weak or non-existent. Display rules’ status as an injunctive norm is likely due to its links to emotional labor research, which tends to sample from service professions (e.g. Diefendorff, Erickson, Grandey, and Dahling, 2011; Allen, Pugh, Grandey, and Groth 2010), where emotion-related expectations are particularly salient (Rafaeli & Sutton, 1987). Looking at the social context of emotional regulation through the lens of descriptive EFC norms in a non-service sample may provide a more complete picture of how employees deal with negative emotions at work. Workgroup EFC norms also advance upon display rules research by providing a fuller description of emotional regulation. The display rules measures describe the expression of affect in terms of acting friendly or excited and the suppression of affect in terms of suppressing boredom and annoyance (Diefendorff, Richard, & Croyle, 2006). EFC behaviors, as framed within cognitive appraisal theories, exhibit a broader range of distinct behavioral categories that illustrate different ways of managing negative emotions. Because they often manifest behaviorally, there is reason to assume EFC behaviors could translate to perceived workplace norms. To examine the empirical relationship between these two constructs, display rules were measured in the present study and compared in an exploratory manner to workgroup EFC norms. Researchers have previously studied the influence of descriptive norms on individual behavior in the area of social influence. Social influence can be divided into two major dimensions: compliance and conformity (Cialdini & Goldstein, 2004). Whereas compliance refers to responses to social requests, conformity refers to altering one’s behavior to be in alignment with others. The influence of workgroup norms for emotion-focused coping represents an expression of conformity in the workplace. Individuals conform to social norms for several reasons, including gaining accurate information about their environment, developing 10 and maintaining social relationships, and preserving a positive self-concept (Cialdini & Goldstein, 2004) Latané’s (1981) social impact theory helps to further explain the mechanisms of social conformity. Latané (1981) posited that individuals conform to social norms as a multiplicative function of the strength, immediacy, and number of other individuals exerting social impact on the target. These perspectives help to explain how the behavioral manifestations of others’ coping may exert influence on an individual over time. The strength of norm sources, meaning the “power, importance, or intensity of a given source to the target”, presumably influences whether the target individual adopts a normative coping behavior (Latané, 1981, p.344). Evidence for this can be in part found in research on the influence of leaders’ expressed emotions among followers (Dasborough, 2006). Individuals may also be influenced by EFC norms through the immediacy of norm sources. That is, proximity of norm sources in time and space may partially determine the influence of the norm. In the context of the workplace, one might assume the behaviors of members of a workgroup that share the same physical workspace with an individual exert more influence than members of the workgroup in a separate physical context. Similarly, the EFC behaviors of others within recent weeks is likely to influence individuals more than the EFC behaviors others engaged in several months prior. The number of norm sources may also determine the influence of workgroup EFC norms. In our context, a single coworker’s EFC behaviors are likely to have less influence than the EFC behaviors of numerous coworkers in the workgroup on average. Taken together, the strength, immediacy, and number of norm sources are what likely create an individual’s perception of the workgroup EFC norms. Further, research on pluralistic ignorance has empirically show 11 individual behavior is more influenced by perceptions of group norms than of actual group norms (e.g. Prentice & Miller, 1993). Though the proposed norms may be considered the aggregation of individual norm perceptions, the most potent influence of norms is assumed to occur at the level of individual perception. A model of these workgroup EFC norms is described as follows. A Proposed Model of Workgroup Norms for Emotion-Focused Coping The proposed model, shown in Figure 1, establishes how a novel construct, workgroup EFC norms, links to both proximal and distal outcomes. Workgroup emotion-focused coping norms are defined as a general tendency for group members to engage in types of behavioral efforts to master, tolerate, or reduce fear, anger, distress, guilt, or sadness in the workplace. The way others have dealt with negative emotions in this context may become salient, making those behaviors seem more effective and viable. These norms constitute a bottom-up level of analyses that is formed through the aggregation of individual norm perceptions. However, the norm’s influence is assumed to have the strongest influence on individual affect, cognition, and behavior when considering that individual’s perception of the norm. Workgroup EFC norms may influence proximal outcomes, such as EFC behaviors in response to negative emotions, and distal, organizationally relevant outcomes, including workgroup satisfaction, physical strain symptoms, person-environment workgroup fit, and workgroup helping intentions. These relationships are further discussed in the sections that follow. Proximal Outcomes: Affective Event Theory and Typical Coping Behaviors Affective events theory provides a means of examining the proximal influence of workgroup EFC norms (Weiss & Cropanzano, 1996). Affective events theory posits events 12 should be treated as units of analysis when studying affect. Events are appraised as either congruent or incongruent with one’s own values and goals (Weiss & Cropanzano, 1996). When events are deemed incongruent, negative affect may be elicited within the individual, prompting the individual to react by some means of coping. The work environment itself and the personal disposition of the individual are assumed to influence whether an event elicits negative affect and the outcomes of that affect. Integrating both cognitive appraisal theory and affective events theory, the present model predicts that perceptions of workgroup EFC norms will influence employee’s average EFC responses to several negatively appraised workplace events. As previously discussed, according to Lazarus’s two part cognitive appraisal theory (1991; 1966), individuals make a secondary appraisal of what type of behavior to engage in as a response to the situation. At this point, the individual may reflect on the surrounding context of the situation. In the workplace, the surrounding social context may include the norms for how employees generally cope with negative emotions. When deciding how to respond to the negative emotions elicited by a goal incongruent event, the individual may look to these social cues to decide how they should best respond. Once the individual makes a secondary appraisal, he or she engages in the selected coping behavior. This theoretical framework fits with Latané’s (1981) social impact theory, in that individual may use the strength, immediacy, and number of norm sources as relevant information when engaging in the secondary appraisal process. The present study looks at three types of emotion-focused coping norms (support seeking, use of humor, and emotional expression), and anticipates links to these same categories of emotion-focused coping behaviors. When describing how they might typically respond to negatively valenced workplace events, I predicted that the employees will on average report emotion-focused coping behaviors that are seen as normative in the workplace. 13 Hypothesis 1: Individual’s perceptions of workgroup norms for emotion-focused coping will predict their self-reported typical emotion-focused coping behaviors averaged across negatively valenced situations. Specifically, a perception of a strong norm for a type of coping will positively predict individual reports of typically coping in that manner. Distal Outcomes The anticipated influence of EFC norm perceptions on individual EFC behavior presumably accrues over time. The literature on coping outcomes provides some indication of what physical and attitudinal workplace outcomes might be linked to typical EFC behaviors. Humor has been linked to a variety of positive outcomes, including increased creativity, increased group cohesion, and decreased tension in stressful situations (Romero & Cruthirds, 2006). Emotional expression has been linked to lowered depressive symptoms among those coping with illness (Stanton & Low, 2012). Further, expressive writing, a common manipulation of emotional expression, has been linked to subjective well-being and health (Pennebaker & Chung, 2011). As mentioned previously, meta-analytic research has shown social support to be weakly linked to negative physical outcomes (Penley et al., 2002). The current study’s hypotheses suggested that coping through the use of humor and through emotional expression will be related to positive physical and attitudinal outcomes within the workplace. Potential relationships between social support seeking behaviors and distal outcomes were analyzed in an exploratory nature. Hypothesis 2: Employees’ typical EFC responses to negative affective events at work relate to distal, organizationally relevant outcomes (physical strain symptoms, personenvironment workgroup fit, workgroup satisfaction, and workgroup helping intentions) such that: 14 a: The extent to which employees typically respond to negative affective events at work through emotional expression will relate positively to distal outcomes. b: The extent to which employees typically respond to negative affective events at work through humor will relate positively to distal outcomes. Behavioral Mediation Perceived EFC norms may also exert an influence on distal outcomes for the individual, as mediated by the individual’s typical EFC behaviors. That is, perceptions of how others cope with negative emotions on average are expected to predict how individuals typically cope with negative emotions at work. As described in the justification for hypothesis two, the way individuals cope in general has been linked to positive and negative distal outcomes. The theorybased hypothesized linkages between perceived EFC norms and typical coping behaviors, and those between typical coping behaviors and distal outcomes, provide the basis for the behavioral mediation of perceived workgroup EFC norms’ influence on individual distal outcomes. Typical coping through social support seeking’s mediation of the workgroup norms for coping through social support seeking and distal outcomes were examined in an exploratory manner. Hypothesis 3: Typical EFC responses (emotional expression and use of humor) will mediate the relationships between perceived workgroup norms (emotional expression and use of humor) and outcomes (workgroup satisfaction, workgroup P-E fit, physical strain, and workgroup helping intentions). Self-Esteem as a Moderator The present study also explored potential moderators of the relationships between EFC norms, typical EFC behaviors, and outcomes. The plasticity hypothesis suggests individuals low in self-esteem will be more likely to be influenced by social cues than individuals with high self15 esteem (Campbell, 1990). For example, past research has shown role conflict is more closely linked to health problems among those with low self-esteem than those with high self-esteem (Ganster & Schaubroeck, 1991). Extending this proposition to the current model, individuals low in self-esteem may be more likely to engage in EFC behaviors that are more normative than individuals that are high in self-esteem. For these reasons, self-esteem was included in the present model as a potential moderator of the relationship between perceived EFC norms and EFC behaviors, such that those low in self-esteem will show stronger links between perceived EFC norms and their related EFC behaviors than those with high self-esteem. Hypothesis 4: The relationships between perceived workgroup EFC norms and EFC behaviors will be moderated by self-esteem, such that the relations between perceived EFC norms and EFC behaviors will have a stronger positive relationship when individuals have low self-esteem. Trait impulsivity as a Moderator Trait impulsivity is another characteristic of individuals that may influence reactions to social environments. Low self-control (i.e., high impulsivity) has been linked to counternormative criminal behavior (Pratt & Cullen, 2000). Experimental research has shown individuals with lowered self-control are more likely to break descriptive and prescriptive social norms (DeBono, Shmueli, & Muraven, 2011). Given past findings linking impulsivity to counter-normative behavior (Pratt & Cullen, 2000; DeBono et al.. 2011), the present study predicted individuals with high trait impulsivity would show weaker links between perceived EFC norms and their related EFC behaviors than those among individuals with low trait-impulsivity. Hypothesis 5: The relationships between perceived EFC norms and EFC behaviors will be moderated by trait impulsivity, such that the relations between perceived EFC norms 16 and EFC behaviors will have a weaker positive relationship when individuals are high on trait impulsivity. Gender as a Moderator Gender was also included in the present study as a moderator between emotional expression EFC behaviors and outcomes. Stanton et al. (2000) reports that for women (and not for males), emotional expression showed a significant positive relationship to hope, and a significant negative relationship with the silencing of self scale, a measure of depression. Therefore, in the current model, emotional expression may be more strongly related to positive outcomes for women than for males. Hypothesis 6: Women will have stronger positive relations between emotional expression EFC behaviors and positive distal outcomes than men. Deep Acting and Surface Acting Last, several constructs from the emotional labor literature were explored in the current study for comparison purposes. Emotional labor has been defined as “the management of feeling to create a publically observable facial and bodily display” (Hochschild, 1983). To perform emotional labor, individuals engage in deep acting and surface acting. These behaviors normally occur when there is dissonance between the emotions felt by an individual and the emotional display required by the context. Deep acting has been defined as when one modifies his or her feelings in order to express the desired emotion, whereas surface acting has been defined when one only regulates his or her emotional expression (Grandey, 2000). Conforming to display rules (injunctive norms) and conforming to EFC norms (descriptive norms) have several similarities and differences. They are similar in that for both types of norms, conformity may involve active, thoughtful engagement, as with deep acting, or 17 passive or superficial participation, as with surface acting. Both display rules and EFC norms involve conforming to a perceived “expectation.” However, there are major substantive differences in these constructs. First, the referent of comparison is different for injunctive and descriptive norms. When conforming to display rules, the individual tries to match his or her emotions to an emotional expression he or she is obligated to show. When conforming to descriptive norms for coping with negative emotions, the individual matches what other people do on average to manage their negative emotions. Second, individuals must “act” to meet the requirements of display rules. With EFC norms, conformity represents a shift in coping behaviors, but does not necessarily mean there is dissonance. For example, in a workplace with high norms for emotional expression, members may freely express their negative emotions at work because other people generally do this at their workplace. This does not mean there is a norm for creating emotions to express as display rules would specify, but rather, when an individual does experience a negative emotion, he or she would cope similar to how others cope with that emotion. In a customer service encounter, employees must display happiness and patience, and if he or she does not, the individual must create a means of expressing these emotions. Finally, coping behaviors serve a different purpose than deep and surface acting. The coping behaviors in the current study focus on alleviating the negative feelings that arise from conflicting external and internal demands. How others typically cope with negative emotions is one source of information that may influence how people alleviate negative feelings. Deep and surface acting, which arguably fit under the broadest definition of coping, are oriented towards the goal of presenting a certain emotion. For these reasons, display rules, norms for deep acting and surface acting, and typical deep acting and surface acting were also measured in the present 18 study. Though theory would suggest display rules, surface acting, and deep acting may be similar but different from coping behaviors and norms, analysis of these measures was approached in a largely exploratory manner. 19 METHOD To help establish the external validity of this study, a non-experimental two-part survey design using a field sample was used to test the proposed hypotheses. Though non-experimental designs are limited in making causal inferences given their lack of random assignment, a survey design allowed us to capture the extent to which these norms exist in actual workplaces. Furthermore, given workgroup social norms develop through numerous stages over time (Tuckman, 1965), it would be challenging to manipulate them in an experimental or quasiexperimental design with sufficient psychological fidelity. For these reasons, a survey design was determined to be most appropriate for the proposed project. Pilot data using an undergraduate sample was gathered to ensure the construct validity of the affective events used as part of the survey. Pilot Study Forty-five undergraduates enrolled in psychology courses at Michigan State University with six months prior work experience were used for a pilot study sample. Informed consent was obtained from all study participants using the consent form in Appendix A. Participants were asked in an online survey the extent to which they might feel negative emotions in provided examples of workplace events. This measure is included in Appendix B. Five of these situations were selected for use in the primary study. Several a priori requirements were set for inclusion in the main study. First, situations had to have average ratings of three or higher on one of the negative emotions listed to be included in the primary study. Second, the group of situations selected had to elicit a diverse range of different types of negative emotions. That is, ideally five situations representing high levels of the five emotions (anger, sadness, fear, distress, and guilt) would serve part of the 20 primary study, so that each negative emotion might be represented. Third, situations that show high levels of multiple negative emotions would be maintained in the primary study, as the goal was to have situations that represent negative emotions in general. However, diversity of negative emotions across the situations was also a priority, so that the situations do not simply represent reactions to a single type of negative emotion (e.g. anger experienced at work). The results of the pilot study are included in Table 3. Looking at the table, anger and distress were most frequently rated highly. Only one situation (situation 11-“rumors of layoff”) was rated over 3.00 on the emotion fear, and only one (situation 8-“mistake caused work for others”) was rated over 3.00 on the emotion guilt. Because the goal was to capture the full range of negative emotions, these two situations were included in the primary study. Sadness was only captured in two situations, situation 7 (“not given reward”) and situation 9 (“something you were working on was discarded”). Because the sadness rating was identical for these two situations, situation 7 was included because it had a higher anger rating. Situation 1 (“colleague took credit for your work”) was selected for the main study, because it had a high rating of anger, but was not conflated with other emotions. Situation 4 (‘given information about a meeting”) was selected because it was rated highly on distress and anger, and had high ratings of fear and guilt relative to other situations. The main study was conducted examining these 5 situations. Primary Study Participants and Procedures The target population for the present study was full-time employees in the United States that regularly interact with other employees in the workplace. An a priori power test of twopredictor multiple regression testing R2 increase at an alpha of 0.05 requires a sample size of 107 participants. Data was collected from 293 participants using two sources: management courses at the Michigan State University, and Amazon.com’s Mechanical Turk. Common method variance 21 is a potential concern with survey designs. By measuring the predictor variable at one point in time and the criterion variables at another time point, common method biases may be partially attenuated (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003.), Therefore, norm data was collected during the first data collection, and proximal and distal outcome data were collected the following week. Measuring norms at an earlier date also aided in the study’s ability to draw stronger inferences about the theorized relations between norms and emotion-focused coping behaviors. Mechanical Turk is an increasingly common method of collecting data for psychological research studies (Conway & Peetz, 2012). Recent research has shown data from Mechanical Turk meets psychometric standards and the quality of data collected is generally unaffected by the low compensation amounts (i.e., as low as $.02 for short surveys) (Buhrmester, Kwang, & Gosling, 2011). On Mechanical Turk, the survey was posted as a study about workplace emotions that was for individuals who are employed full-time. Initially, participants were paid $.25 to complete the initial survey, and $.75 to complete a second survey one week later. After 30 participants’ data was collected, the incentive was increased to $1.00 for the initial survey and $3.00 for the second survey to attract more participants and decrease attrition on the second survey. Whereas the first survey was solicited to any Mechanical Turk worker with full-time employment, the second survey was only available to participants that had participated in the first survey. Participants that failed attention checks (e.g. “For quality control, please select ‘strongly agree”) and those that indicated in the demographics section that they were less than ¾ time employed were not included in the data. Surveys were linked using the last three digits of the participant’s phone number, and his or her first and last initial. 96 participants from 22 Mechanical Turk participated in both surveys, 77 participants participated in only the first survey, and 39 participants participated in only the second survey. Participants from the management courses were in either an undergraduate course or an executive MBA course. Participants were given course credit for participation in either survey. If students in the class did not have a full-time job currently, they were allowed to have a relative or friend fill out the survey for course credit. Data was connected using an identifier based on the last 3 digits of the student’s ID number. 53 participants were linked to both surveys, 19 participants participated in only the first survey, and 9 participants participated in only the second survey. It was only possible to link the two-part data for 149 of these participants. There are several potential reasons for this missing data. First, management students had the option of participating in either survey or both. Thus, many likely opted to complete one survey but not the other. Second, if management students were not employed full time, they could ask a friend or family member to take the study for them. Because the identifier required knowing the student’s ID number, friend or family member participants may have taken the survey but did not have immediate access to the student’s ID at the time, prohibiting the linkage to the subsequent survey. Further, management students may have only asked a friend or family member to do one survey and not the other, as these surveys were likely completed as favors for the students. Third, in both management and Mechanical Turk samples, many participants filled out the identifier incorrectly or inconsistently (e.g., incorrect amount of numbers, provided their Mechanical Turk Worker ID instead, etc.), making it impossible to link their data between time points. There were several instances in the Mechanical Turk data where identifiers had matching first and last initials, but the phone numbers differed. It may have been that participants entered different 23 phone numbers for each survey (e.g. intermixing home, work, and cell phone numbers.) It was impossible to verify whether or not this happened, which required the elimination of both sets of responses. Last, the attrition rate was much larger among the Mechanical Turk sample that was paid $0.75 to complete the second survey (approximately 70% drop-out rate) than when participants were paid $3.00 (less than 10% drop-out rate.) Whereas most participants were willing to log into Mechanical Turk a week later to take the second survey for $3.00, the $0.75 incentive may not have been attractive enough for continued participation. Case-wise deletion is generally not encouraged; however, alternative methods of data imputation were not desirable with such a large amount of attrition. Further, the relationships of interest could not be examined with single survey data. Given there were no significant differences between participants with complete (i.e., individuals with both time points) and incomplete data (i.e., individuals with only time 1 or time 2 data) on demographics, EFC behaviors and EFC norms, analyses were conducted using the 149 participants with linked data. The theoretical premises of this study suggested that it is important to examine the EFC of employees in industries including but not limited to service industries. To capture what percentage of the current sample was from the service industry, self-reported industries and jobs were coded as service or non-service. Job and industry codings were used, as some participants worked in service industries (e.g., hospitality), but did not have service jobs (e.g., CFO). 37.6% reported working in service industries (e.g. education, healthcare, retail) whereas 62.4% reported working in non-service industries (e.g. manufacturing, finance). However, the job codings showed even smaller numbers of participants working in actual service positions (24.8%), whereas the majority worked in non-service jobs (75.2%). 24 Consent for the first survey was obtained using the form in Appendix C (MSU sample versions) and Appendix D (Mechanical Turk version). Consent forms for the second survey can be found in Appendix E (MSU sample version) and Appendix F (Mechanical Turk version). Primary Study Measures In general, shorter measures were given preference when selecting measures for this study, to ensure a survey length that maximizes participant engagement. Rating scales were varied across the measures to minimize common method variance. Primary Study Survey 1 Self-Esteem. Self-Esteem was measured using the Rosenberg Self-Esteem Survey (RSES) (Rosenberg, 1965). The RSES is a widely used measure of self-esteem (Blascovich, & Tomaka, 1991). Items were rated on a scale of 1 (Not at all True of Me) to 4 (True of Me to a Great Extent). A mean score of these items was computed. The inter-item reliability for this scale was α=.80. This measure can be found in Appendix G. Trait Impulsivity. Trait Impulsivity was measured using the 30 item Barratt Impulsiveness Scale (BIS; Barratt, 1959). The BIS is a well-established measure of impulsiveness (Stanford et al., 2009). Items were rated on a scale of 1 (Rarely/Never) to 4 (Almost Always/Always), and were averaged together into an overall mean score. The inter-item reliability for this scale was α=.81. This measure can be found in Appendix H. Workgroup Emotion Focused Coping Norms. To measure norms for emotion focused coping, individuals were asked to rate the extent to which they saw coping behaviors as typical within their workplace. The instructions read, “Imagine how the people you work with generally deal with negative feelings, such as sadness, anger, fear, distress, or guilt. Rate the following items based on the extent to which you think individuals in your workgroup would use them to respond to negative feelings.” The coping items were selected from the measures below, because 25 they captured differing aspects of emotion-focused coping and were behavioral in nature. Agreement with items were measured using a 5-point Likert scale, and a mean score was computed for each norm below. The full measure can be found in Appendix I. Social support seeking norms (instrumental and emotion-focused). Support seeking was measured using the COPE’s support seeking-instrumental (Carver et al., 1989; α =.75) and support seeking-emotional (Carver et al., 1989, α=.85) dimensions (items 1-4 and 5-8 in Appendix I, respectively). The inter-item reliability for the instrumental scale was α=.76; the inter-item reliability for the emotional scale was α=.85. Coping through humor norms. Four items were modified based on Martin and Leftcourt’s (1983) measure for coping through humor (items 9-12; Appendix I). A review of the literature showed this scale is the only popular measure of coping through humor that consists of more than 1 item. Martin and Leftcourt’s (1983) original items were more dispositional than situational. For example, items referred to “having a sense of humor.” Second, the items do not focus on behaviors. For example, “finding something funny” is a cognitive activity, and not a behavior like “making a joke.” The revised scale was more behavioral and situation based. The inter-item reliability for this scale was α=.89. Emotional expression norms. Four items from Stanton et al.’s (2000) emotional expression measure were used to capture the behaviors involved in letting feelings come out freely (items 13-16; Appendix I). Stanton et al. (2000) note this measure was developed to capture emotion-focused coping behaviors that are less pathological and neurotic than items of past measures (e.g. venting; Carver et al., 1989). In comparing emotional expression measures (Stanton et al., 2000) and emotional suppression measures (Gross & John, 2003), it appeared that these two constructs could be 26 capturing polar ends of the same continuum. For example, Gross and John’s (2003) suppression items refer to “not expressing” emotions, indicating that suppression is a lack of expression. To address this issue, both Stanton et al.’s (2000) measure and three items from Gross and John’s (2003) measure (items 17-19; Appendix I) were included in the study. Gross and John’s item “When I am feeling positive emotions, I am careful not to express them” was not included, because the current study focuses on only the expression of negative emotions. A factor analysis was conducted to test the dimensionality of these items. A single factor solution seemed to best match the data, with the first factor accounting for 61% of the variance. Thus, emotional suppression norm scores were reverse scored and averaged into the emotional expression norms measure. The inter-item reliability of the combined items was α=.90. Surface and deep acting norms. To compare norms for emotion-focused coping with norms for surface and deep acting, items from Brotheridge and Lee’s (2003) deep and surface acting measure were included among the norm items. Looking at the content of the items, the surface acting items (items 20-22; Appendix I) on “resisting” and “hiding” true feelings are captured by the general construct of emotional expression/suppression. “Pretending to have emotions that I don’t really have” emphasizes the need to present an emotion, and not actually how the negative feelings are alleviated. The deep acting items (items 23-25; Appendix G) make clear that the referent for these behaviors is not what people do on average, but instead, what emotional displays are “needed” or “must be shown.” It is important to note that the stem for this measure was that of a descriptive norm (i.e., what others do on average), despite these items typically referring to injunctive contexts (i.e., display rules). The inter-item reliability for the surface acting norms scale was α=.71; the inter-item reliability for the deep acting norms scale was α=.83. 27 Display rules. To examine the distinction between display rules and the focal construct, norms for emotion-focused coping, Best, Downey, and Jones’ (1997) measure of display rules was included in the survey (items 1-3; Appendix J). This specific measure has been used in prior published research (Brotheridge & Grandey, 2002; Diefendorff, Erickson, Grandey, & Dahling, 2011). Items were rated on a scale of 1 (not at all required) to 5 (always required), and a mean of the items was computed. The inter-item reliability for this scale was α=.82. Demographics. The demographic questions included in this survey are listed in Appendix K. Primary Study Survey 2 The second survey was administered approximately one week after the first survey. Typical EFC Behavior. Based on findings from the pilot study (Appendix B), five situations were selected as representative of typical workplace events that elicit a variety of negative emotions. Participants rated whether or not the actual event had ever occurred to them at their job, and if not, were asked to rate the item according to how they think they might cope with that situation. As a manipulation check, participants were asked to rate the extent to which they felt or would feel anger, sadness, fear, distress, and guilt in the provided situation. Ratings for each negative emotion were on a scale of 1 (not at all) to 5 (extremely). Participants reported how they generally respond to each type of situation by rating the coping items found in the workgroup norms measure (i.e., social support seeking, coping through humor, emotional expression, emotional suppression, surface acting, and deep acting; Appendix L) on a scale of 1 (never) to 5 (almost always). Again, a factor analysis was conducted to examine whether typical coping through emotional expression and emotional suppression loaded onto a common factor. A single factor 28 solution seemed to best describe the data, as the first factor accounted for 47% of the variance in eigenvalues, and the second dropped down to 10%. With typical emotional expression and reverse-scored typical emotional suppression items from all situations combined, the average inter-item correlation was r=.43, and the alpha coefficient was α=.96. These inter-item correlations and alphas were as high or higher when examined within each situation (For situation 1 through 5; average inter-item correlation respectively, r=.58, r=.59, r=.71, r=.72, and r= .70; reliabilities, α=.91, α=.91, α=.94, α=.95, and α=.94.) For these reasons, the suppression items were reverse-scored and averaged as part of the overall emotional expression score. The resulting inter-item reliabilities for typical behaviors were as follows: social support seeking-instrumental, α=.90; social support seeking-emotional, α=.96; use of humor, α=.94; emotional expression, α=.96; surface acting, α=.93; and deep acting, α=.96. Because all reliabilities with these situations collapsed were over .90, it was deemed appropriate to calculate overall scores that averaged all coping items within a dimension across the 5 situations (e.g., social support seeking-emotional items in situation 1, 2, 3, 4, and 5), and use these for analyses. Physical Strain. Physical strain was measured using a 13 item measure from Spector and Jex (1998; α=.87). These items are included in Appendix M. The instructions asked participants to rate the extent to which the experienced symptoms of physical stain within the past 3 months, and items were rating using a frequency scale of 1 (less than once per month or never) to 5 (several times per day.) Items were averaged into a total mean score. The inter-item reliability for this scale was α=.83. Workgroup Helping Intentions. Workgroup Helping Intentions were measured using modifications of 6 items from Van Dyne and LePine’s (1998) helping measure (α=.85; Van Dyne & LePine, 1998). While these items were intended to measure behaviors as rated by the 29 self, peers, or supervisors, the wording was adjusted to capture future-oriented self-ratings. Items are included in Appendix N, and were averaged together for an overall mean score. The interitem reliability for this scale was α=.92. Workgroup Satisfaction. Workgroup satisfaction included items from the people on your present job dimensions of the Job Descriptive Index (Roznowski, 1989). This measure is commonly used to measure job satisfaction by measuring how individuals feel across several dimensions. The “people on your present job” subdimension measures the extent an individual is satisfied with his or her colleagues. Whereas the typical JDI scoring uses an ordinal scale, a 5point Likert scale is used in this study for survey consistency. Reverse scored items were rescored and items were averaged to create an overall scale score. Items are included in Appendix O. The inter-item reliability for this scale was α=.94. Workgroup Person-Environment Fit. A six item person-environment fit measure was used in the present study and can be found in Appendix P. Items were based off of Seong and Kristof-Brown’s (2012) values-based and personality-based team fit measures (respectively, α=.0.94; α=0.95; Seong et al., 2012). Items were rated 1 (disagree), 2 (undecided), or 3 (agree). Items were averaged together for a total score. The inter-item reliability for this scale was α=92. 30 RESULTS Sample Descriptives A description of the sample can be found in Table 4. The sample was mostly mid-career and White. Of particular interest is proximity to colleagues. The majority of participants worked in close proximity to coworkers, indicating frequent in-person interaction with workgroup members. The jobs represented in the sample were drawn from a broader array of industries than the service industry samples in other emotional regulation studies (Diefendorff et al. 2011; Allen et al., 2010). The study included accountants, managers, IT analysts, and financial analysts, from banking, automotive, and insurance industries. Factor Analysis of Descriptive and Injunctive Norm Items Several exploratory factor analyses were conducted to examine the underlying structure of the data, and to better understand how measures included for comparison (surface acting, deep acting, and display rules) related to measures of interest (workgroup EFC norms and EFC behaviors). Note that surface acting and deep acting norm items were measured using the same item stem as EFC norms (i.e., what do workgroup members do on average), meaning surface acting norms and deep acting norms were assessed as potential descriptive norms. It is also important to recall display rules (injunctive norms) were measured in addition to descriptive surface acting norms and descriptive deep acting norms; meaning these three scales are not one and the same. In interpreting the results of the following factor analyses, it should be noted that although the present sample meets more lenient standards for number of participants when conducting a factor analysis (e.g. sample size of 100 or 5 to 1 participant: variable ratio; cf. MacCallum, Widaman, Zhang, & Hong, 1999), it does not meet more stringent standards (e.g. 10 31 cases per item included; Nunnally, 1978, p.276) EFC norms, surface and deep acting norms, and display rules were entered into a maximum likelihood estimation factor analysis. The rotated factor matrix can be found on Table 5. Based on the variance explained by each factor and the scree plot, a seven-factor solution was adopted. Constructs of interest were clearly distinguished with this factor solution. However, there are several aspects of the factor structure that are worth noting. First, despite past research indicating the instrumental and emotional subscales of seeking social support are highly correlated (r=.69, Carver et al., 1989), these norm items mapped on different factors. Second, display rules mapped onto a single factor, distinct from workgroup EFC norms. Third, as expected, emotional expression norm items and emotional suppression norm items mapped on to the same factor. Fourth, descriptive norms for deep acting and surface acting were measured in an exploratory manner to examine how they relate to the other norms. Descriptive deep acting norms mapped onto a factor separate from EFC norms. Unexpectedly, two of the three items from descriptive surface acting norms mapped onto the first factor with emotional expression and suppression norm items. These surface acting norm items emphasize emotional suppression of true feelings (i.e., “Resist expressing their true feelings”; “Hide their true feelings about a situation”), which arguably creates overlap with emotional suppression items (e.g., “Make sure not to express the negative emotions they are feeling.) Fifth, an emotional-focused social support norm item (“talk to someone about how they feel”) mapped onto a factor with emotional expression norm items. Although not originally expected, emotional expression norm items and emotion-focused social support norm items may have substantive overlap. One way individuals may engage in emotional expression may be talking to others about the negative feelings they experience. 32 Factor Analyses of Typical EFC Behaviors In examining the factor structure of typical coping behaviors, it was not possible to examine all typical coping items together due to the large number of items (i.e., 5 situations with 25 items; 125 items total). Consequently, factor analyses were conducted at the level of situation. These factor analyses can be found in Tables 6 through 10. The factor pattern was not identical across all situations. Situation 1 (“colleague takes credit for your work”, a high anger situation) resulted in a 6-factor solution (Table 6) whereas Situations 2 through 5 produced 5-factor solutions (Table 7-10). Additionally, in two of the situations, typical coping through instrumental and emotion-focused social support seeking items loaded on the same factor (Situation 2, Table 7, “colleague makes you late to meeting”, high anger and distress ; Situation 4, Table 9, “your mistake causes extra work for others”, high distress and guilt ). Yet the factor structures are similar in a variety of ways. First, typical coping through the use of humor consistently loaded onto a single factor. Also, the first item in the typical use of humor scale (which was reverse scored) tended to load on the humor factor more weakly than the other items. Second, typical emotional expression behavior items frequently loaded on a factor with typical emotional social support seeking behaviors. Third, deep acting behaviors consistently mapped onto a single factor. Fourth, as in the norms factor analysis, typical emotional expression behavior, typical emotional suppression behavior, and typical surface acting behavior items tended to emerge on the same factor. Although these norm and typical behavior factor analyses and correlations seem to suggest a heavy overlap between emotional expression and surface acting (norms, r= -.69, typical behaviors, r= -.77), surface acting was analyzed separately from emotion expression for the sake of comparison with prior surface acting research. 33 Revision of Humor and Social Support Scales Based on the factor analysis examining EFC norms and those examining typical EFC behaviors, several revisions were made within subsequent analyses. First, the reverse scored humor item (“find it hard to laugh it off with other people”) showed weaker loadings than the others, in both norms and typical coping behavior. The item had had much poorer inter-item correlations (i.e., generally between r=.30 and r=0) than those found between other items (i.e., generally between r=.30 and r=.60) for both norms and coping behaviors. Additionally, interitem reliability increased when this item was excluded (typical behaviors, from α=.91 to α=.94; norms, from α=.83 to α=.89). For these reasons, the reverse scored item was excluded from the EFC norm and typical EFC measures in future analyses. Second, the factor analyses showed instrumental and emotion-focused social support seeking did not overlap as strongly as initially anticipated. Further, the correlations between measures of these as EFC norms and as typical EFC behaviors (respectively, r=.17; r=.49) were weaker than those in past research (r=.69, Carver et al., 1989). For these reasons, instrumental and emotion-focused support seeking were analyzed as separate constructs in subsequent analyses. Particular interest for this study was with emotion-focused social support seeking, given it is an emotion-focused and not problem-focused (as with instrumental support seeking.) Descriptive Statistics and Intercorrelations of Separate Situations To better understand the influence of each situation on variables of interest, and to more definitively assess whether the situation data should be collapsed, several other descriptives and intercorrelations were generated. First, in Table 11, the descriptives for the manipulation check are shown. The findings show that the situations were rated as eliciting emotions similar to those rated in the pilot study. Again, situation 1 (“colleague has taken credit for your work”) elicits 34 feelings of anger, whereas situation 2 (“given incorrect information about a meeting”) elicits anger and distress. In the pilot study, situation 3 (“a reward you were expecting was not given to you”) generated high levels of anger and sadness; in the primary study, this event also elicited high levels of distress. Similar to the pilot study, situation 4 (“your mistake resulted in extra work for others”) elicited high levels of guilt. Situation 5 (“you hear rumors of layoffs”) again was rated as generating high levels of fear and guilt. Thus, as expected from the pilot, the situations encompassed high levels of the five emotions of interest. A separate analysis of the manipulation check was run comparing situational data for individuals who did and did not mark that they had previously experienced a similar event at work. The manipulation check was nearly identical, indicating that, for any particular situation, participants who had experienced the situation responded similarly on manipulation check items to participants who had not. It was also of interest the extent to which measures of behavior responses to each situation exhibit similar reliabilities, similar within-situation correlations, and similar correlations with norms and outcomes. Table 12 shows within-situation behavior descriptives, reliabilities, intercorrelations with other behavioral dimensions, and intercorrelations with norms. Means tended to hover around the middle of the scale for all dimensions, with standard deviations close to 1.00. Reliabilities were all over .80. Social Support Seeking-Instrumental and Social Support Seeking-Emotional generally showed positive relationships with other behavioral dimensions, with the exception of surface acting, with which it showed a negative relationship. Humor and Social Support Seeking-Emotional had a null relationship in situation 1 (“coworker takes credit for your work”), whereas Social Support Seeking-Emotion had an especially strong relationship with emotional expression (r=.61). This may be because situation 35 1 may be conceived as an incident of betrayal, wherein emotional social support and emotional expression involve commiseration rather than finding humor in the situation. Humor otherwise tended to show small relationships with emotional expression behaviors. Humor showed no clear relationship with surface acting, showing these constructs are distinct from one another. Surface acting had strong negative relationships with emotional expression, showing surface acting may be a behavior opposite of emotional expression. Deep acting behaviors tended to show positive relationships with EFC behaviors (i.e., social support seeking, humor, emotional expression.) This may be because deep acting behaviors reflect a positive commitment to workgroup norms, which may also explain why individuals feel free to express their emotions with coworkers and ask them for help. Looking at the relationships of situational behaviors with norms in Table 12, the normtypical behavior relationships of interest (in bold) were of similar sizes across situations. Use of humor coping behaviors were not as strongly related to Use of Humor Norms in Situation 2 (“given incorrect information about a meeting”) and 4 (“your mistake caused extra work for others”). The smaller relationships may be due to the fact that these situations reflect tensions between coworkers, rather than more collectively negative experiences in the workplace (such as, “you hear rumors of layoffs”). In general, the relationships between behaviors within situations, and the relationships between norms and typical behaviors supported the collapsing of behavioral dimensions (EFC behaviors, surface acting, and deep acting) across situations. Table 13 examined the situational data’s correlations with outcomes and other variables (gender, race, work experience, physical proximity, and position tenure). Very few of these relationships were significant. Gender was consistently related to emotional social support 36 seeking, with women being more likely to engage in this behavior. This finding is consistent with past findings (Carver et al., 1989). Counter to past research (Stanton et al., 2000), the relationship between gender and emotional expression was inconsistent and at times, null. Physical proximity showed significant negative relationships with surface acting and deep acting in situation 5 (“you hear rumors of layoffs”), meaning individuals who have less frequent inperson interactions with co-workers would engage in more surface and deep acting in this situation. One potential explanation for this is that individuals who work remotely may feel they are in greater danger when a company is looking to lay-off employees, and thus, must engage in more impression management to highlight what effective performers they are. Deep acting tended to show a significant positive relationship with position tenure. This may be because employees who have stayed in their position for long periods of time are more highly socialized into their workplace. Individuals who are more embedded in their roles may feel more motivation to cultivate the affective behaviors they are obligated to show for their job. Impulsivity showed significant, negative relationships with instrumental social support seeking behaviors in situation 3 (“you were not given a reward you were expecting”) and situation 5 (“you hear rumors of layoffs.”) This suggests less impulsive individuals may be more likely to seek out advice from others in situations that involve sense-making of an organization’s actions. The relationships between situation behaviors and outcomes showed general consistency across situations, supporting the averaging of typical behaviors (EFC behaviors, surface and deep acting) across situations. Thus, analyses proceeded using averages of typical behavior across these 5 situations. Descriptive Statistics and Intercorrelations of Collapsed Situations 37 Descriptive statistics, scale reliabilities, and intercorrelations for all study measures can be found in Table 14. Many of the unexpectedly large relationships amongst EFC norm dimensions, and amongst EFC behavior dimensions paralleled factor analytic and situation-based correlation findings. For example, within both EFC norms and typical EFC behaviors, emotional expression showed strong positive associations with emotional social support seeking, and strong negative associations with surface acting. The relationships between emotional expression coping and both social support seeking dimensions of typical coping (instrumental-focused social support, r=.37; emotion-focused social support, r=.57) were similar in size to relationships in prior research (composite of instrumental and emotion-focused, r=.44; Danoff-Burg, Prelow, & Swenson, 2004). The relationship between perceived display rules and typical surface acting behaviors (r=.21) was similar to the somewhat inconsistent relationships found in past research (r=.15, Grosserand & Diefendorff, 2005; r=.38, Diefendorff et al., 2011). The relationship between display rules and typical deep acting behavior (r=.11) was smaller than found in other studies (r=.16, Grosserand & Diefendorff, 2005; r=.20, Diefendorff et al., 2011). Typical deep and surface acting behavior showed a weak, positive relations (r=.10), which lies between past mixed findings (r=.03, Grosserand & Diefendorff, 2005; r=.26, Diefendorff et al., 2011). Proximity to colleagues showed a significant positive relationship with norms for emotional social support seeking, norms for deep acting, and typical deep acting behavior. One potential explanation for the positive relationship between proximity and emotional support seeking is that employees who work in the same space may feel closer to one another and thus, may exchange more emotional support to one another. The positive relationships between proximity and deep acting norms, as well as the relationship between proximity and typical deep acting behavior were also surprising. Deep acting may have far less value to employees who do 38 not actually have to interface with others in-person. Emotional regulation does not actually apply to individuals working remotely, given no one sees the emotional responses of remote employees. In-person interactions require individuals to know and display what emotions are most appropriate for the context. Control Variables Gender, race, and sample source (e.g., management course vs. Mechanical Turk) were considered as control variables. All the regression analyses that follow were conducted with and without the control variables entered as an initial step. The main effect of gender, race, or sample source did not change the significance or direction of any the regression findings, with one exception noted below. Because the results were not substantively altered when controls were entered, the analyses that follow do not include gender, race, or sample source as a control. To test if compensation amount had an influence on relationships of interest in the Mechanical Turk data, all regressions from the hypotheses were conducted on the data with and without a dummy-coded payment variable entered in as a control. The directionality and significance of the findings did not change when the regressions controlled for compensation amount. Thus, compensation was also not controlled for in the analyses. Hypothesis 1 Hypothesis 1 anticipated that perceptions of strong emotion-focused coping norms would predict the extent to which individuals report engaging in corresponding emotion-focused coping behaviors during negative events. The results of these regressions are displayed in Table 15. Each of the coping norms significantly predicted their corresponding coping behaviors. Thus, Hypothesis 1 was supported by the data. 39 Surface acting, deep acting, and display rules were examined in an exploratory manner. As can be seen in Table 15, descriptive norms for surface acting significantly predicted typical surface acting behaviors, but deep acting descriptive norms did not significantly predict typical deep acting behavior. Display rules’ prediction of coping behaviors, deep acting, and surface acting was also explored. The results of these analyses can be found in Table 16. Display rules did not predict typical coping behaviors or deep acting behavior, but they did significantly predict surface acting behavior. When gender and race were entered as control variables, display rules also significantly predicted emotional support seeking (B=.13, SE= .06, β= .17, p=.04). Hypothesis 2 Hypothesis 2 predicted employees that typically engage in a) coping through emotional expression and b) coping through the use of humor during negative events would experience positive distal workplace outcomes (lowered physical strain, higher P-E fit, higher workgroup satisfaction, and higher workgroup helping intentions). Social support seeking, surface acting, and deep acting behaviors were examined as predictors of these outcomes in an exploratory manner. Emotional social support seeking, instrumental social support seeking, coping through the use of humor, and coping through emotional expression did not significantly predict any of the outcomes (Table 17). Thus, hypothesis 2 was not supported. Surface acting and deep acting were also explored as potential predictors of these workplace outcomes (respectively, Table 18). Surface acting had a significant positive association with physical strain, as did deep acting (p<.05). For Hypothesis 1 and 2, supplemental analyses were conducted comparing situational data for individuals who did and did not mark that they had previously experienced a similar event at work. To do this, events that were not previously experienced were removed from the 40 data, and scale scores were recalculated using only events that were experienced previously. Regression analyses for hypotheses 1 and 2 were rerun. The direction of the resulting statistics and the significance levels were not changed, indicating that again, for any particular situation, participants who had experienced the situation responded similarly on typical EFC behavior items to participants who had not. Hypothesis 3 Hypothesis 3 predicted that typical EFC responses would mediate the relationships between perceived workgroup norms and outcomes. Because the relationships between typical EFC behaviors and outcomes were all non-significant, hypothesis 3 could not be tested. Although descriptive norms for surface acting and display rules both significantly predicted surface acting, and surface acting significantly predicted physical strain, tests of the direct influence of norms on outcomes (Table 19 and 20) showed surface acting norms and display rules did not significantly predict strain. Thus, typical surface acting behavior was not a mediator of normative influence on physical strain. The only significant relationship between norms and outcomes was with display rules, which significantly predicted workgroup fit and workgroup satisfaction (Table 20); however, there was no explanatory mechanism to support testing mediation of this effect. Hypothesis 4 Hypothesis 4 predicted that relationships between perceived workgroup EFC norms and EFC behaviors would be moderated by self-esteem, such that the relations between perceived EFC norms and EFC behaviors would have a stronger positive relationship when individuals have low self-esteem. In a hierarchical regression, self-esteem and norms were entered as a first step and the interaction term was entered as a second step. As can be seen in Table 21, none of 41 the norm-coping behavior relationships were significantly moderated by self-esteem. Hypothesis 4 was not supported by the data. Table 22 also shows tests of self-esteem’s moderation of the relationships between surface acting norms and surface acting behavior, deep acting norms and deep acting behavior, display rules and surface acting behavior, and display rules and deep acting behavior. Selfesteem’s moderation was non-significant for each analysis. Hypothesis 5 Hypothesis 5 predicted that the relationships between perceived workgroup EFC norms and EFC behaviors would be moderated by trait impulsivity, such that the relations between perceived EFC norms and EFC behaviors will have a weaker positive relationship when individuals are high on trait impulsivity. Again, using an interaction term in hierarchical regressions, impulsivity only significantly moderated the relationship between norms for emotion-focused social support seeking and emotion-focused social support seeking behaviors (Table 23; ΔR2 = .03, F(3, 145) = 4.39, p <.001). A graph of this interaction can be found in Figure 2. The figure indicates this interaction was in the expected direction. That is, individuals with low impulsivity appear to be more influenced by perceptions of norms than individuals with high impulsivity. Thus, hypothesis 5 received partial support. Table 24 shows tests of impulsivity’s moderation on the surface acting, deep acting, and display rules regression tested in Table 22. Impulsivity also did not significantly moderate any of these relationships. Hypothesis 6 Hypothesis 6 predicted that women would have stronger positive relations between emotional expression behaviors and positive distal outcomes than men. As can be seen in Table 42 25, when an interaction term was entered into a hierarchical regression, gender did not significantly moderate any of the relationships between emotional expression and distal outcomes. Thus, Hypothesis 6 was not supported by the data. Table 21 provides an overview of all hypotheses and whether they were supported. Exploratory Analysis: Proximity as a Moderator of Norm-Typical Behavior Relations Latané’s (1981) social impact theory suggests that physical proximity leads to greater norm conformity. Physical proximity data was collected, making it possible to test potential moderation of the norm-behavior relationship. Tests of physical proximity’s moderation of EFC norms and behaviors are included in Table 27. Counter to what Latané’s (1981) social impact theory suggests, physical proximity did not moderate these relationships. Table 28 also shows a test of physical proximity’s moderation on the surface acting norm-behavior relationship, deep acting norm-behavior relationship, display rules-surface acting relationship, and display rulesdeep acting relationship. For all tests, the effect of physical proximity was non-significant. 43 DISCUSSION Findings from the present study indicate that employees may cope with negative emotions in ways that they perceive to be normative, yet these EFC behaviors have little influence on organizationally important outcomes. The linkages between typical EFC behaviors and employee outcomes (i.e., workgroup satisfaction, workgroup P-E fit, workgroup helping intentions, and physical strain) were not statistically significant. Surprisingly, self-esteem and trait impulsivity had little influence on the relationships between EFC norms and EFC behaviors. Women also did not appear to experience increased positive outcomes from emotionally expressive coping in comparison to men. Overall, this study was able to capture information about norms and the emotional life of workers that expands beyond display rules, surface acting, and deep acting. It also expands our understanding of the relationships between the workplace behaviors that focus on the social emotional displays (i.e., surface acting, deep acting, and display rules), and the workplace behaviors that focus on the alleviation of negative emotions (i.e., EFC behaviors). Study contributions are more specifically addressed in the sections that follow. Emotions in Non-service Settings Previously research on emotional labor has tended to utilize samples of individuals in service jobs (Diefendorff et al., 2011; Allen et al., 2010). Yet recent research has shown that some levels of surface acting and deep acting behaviors are reported by individuals who never or rarely interact with customers, and these behaviors have similar antecedents and outcomes across industries (Hunter, Rubino, Perry, & Penney, 2013). The present study’s sample consisted of largely non-service jobs and industries. Participants reported engaging in surface and deep acting in response to situations that elicit negative emotions. Additionally, participants reported that others in their workgroup to some extent engage in surface acting and deep acting behaviors. 44 Interestingly, the present sample showed typical surface and deep acting behaviors predicted physical strain, but not other outcomes (workgroup helping intentions, workgroup satisfaction, or perceptions of fit.) In general, the relationships between surface acting, deep acting, and display rules matched past research (Grosserand & Diefendorff, 2005; Diefendorff et al., 2011) This presents a point for future research: although individuals in non-service jobs report hiding emotions they are required not to show, when do these encounters with “emotional obligations” occur? The finding that self-reported workgroup attitudes do not significantly suffer when surface or deep acting are reported may indicate that surface and deep behaviors occur when group members interface with individuals outside of the group (e.g., with a boss, with other groups in the organization). Or, it may be that while self-reported workgroup attitudes do not suffer among individuals who report surface or deep acting, others in the workgroup experience negative consequences. Further, the negative events selected for this study may be more infrequent than the negative service encounters that service employees handle regularly, which could explain the lack of linkages between EFC, surface, and deep acting behaviors and more distal outcomes. Distinguishing EFC, Surface and Deep Acting, and Display Rules A core assumption of this study was that EFC norms differ from display rules, and that EFC behaviors differ from deep acting and surface acting behaviors. Looking first at the norm perceptions, factor analytic findings support the notion that display rules are in fact distinct from EFC norms. These same factor analytic findings also indicate that descriptive norms for surface and deep acting are distinct from display rules. At first blush, one might expect that display rules and descriptive surface and deep acting norms would be positively correlated, as perceptions of high display rule requirements might be associated with perceptions of high levels of deep and 45 surface acting norms to meet environmental demands. However the distinction between display rules and descriptive surface and deep acting norms is likely due to the fact that display rules are an injunctive norm, whereas surface and deep acting norms are descriptive norms. That is, display rules reflect perceptions of what people are obligated to do, whereas surface acting and deep acting norms reflect perceptions of what people do on average. This is further reflected in the low correlations between display rules and deep acting (r = -.13), and display rules and surface acting (r = .22,). Also, display rules may be distinct from descriptive deep acting norms in part because display rules are negatively valenced (e.g., people generally hide emotions), whereas deep acting norms were positive valenced (e.g. people really try to show the emotions they are supposed to exhibit for their job). The relationships between behavioral dimensions were more complex. Whereas deep acting behavior emerged on its own factor in each situation, surface acting continually emerged on the same factor as emotional expression and suppression items. Emotional social support seeking also tended to map onto a factor with emotional expression and suppression items. These relationships seem to indicate an overall theme of the open sharing of emotions, a concept that may unite EFC dimensions and surface acting behaviors. Work contexts that inhibit emotional expression and discourage seeking emotional support from others may lead workers to engage in surface acting. The current study’s literature showed these constructs have not been previously linked in the literature. This presents several implications for future research. First, researchers exploring coping should be aware of the high relationships between Social Support Seeking-Emotional items in Carver et al.’s (1989) COPE measure and Stanton et al.’s (2000) measure of emotional expression. Second, although there are likely construct overlaps between emotional expression, 46 emotional social support seeking, and surface acting, items in these three measures seem to lack specific context. For example, are individuals seeking emotional social support from their boss or mentor? Or from the coworker next to them? Similar measures have been used to measure social support (van Daalen, Willemsen, &Sanders, 2006). Who are these emotional confidants and when are they approached for support? Do individuals without confidants in the workplace feel they need to engage in more surface acting? The present study’s measures of typical behaviors provided an antecedent context, but the EFC behavioral items and surface acting items did not provide much information about the subsequent coping context. Future research on emotional expression, support seeking, and surface acting should include items with more information about the context to assess when and why people engage in these behaviors. Last, the notion that surface acting may be on the same continuum as emotional expression supports Stanton et al.’s (2000) suggestion that emotional expression should be viewed as a healthy behavior. Given surface acting was linked to physical strain in the present study; it suggests that a lack of open emotional expression may in fact be unhealthy for individuals in the workplace. However, this suggestion is limited by the present study’s finding that there were no significant relationships linking emotional expression behaviors and emotional support seeking behaviors with positive distal outcomes. Situational Differences and the Distinction between Social Support Seeking-Emotional and Social Support Seeking-Instrumental The factor analyses across the different situations were not identical; thus, the content of each situation was examined to investigate whether emotional reactions or other situational factors could account for these differences. Situation 1 (“colleague takes credit for your work”), had a 6 factor solution, whereas all other situations had 5 factor solutions. This situation was 47 chosen because it was rated as eliciting high levels of anger. Yet anger was an emotion common across situations, making it unlikely the emotions experienced in this situation explain the differences in the factor structure. What seems more likely is participants were required to repeatedly answer similar items, which may have caused fatigue and boredom. After the first situation, participants may have responded with less granularity than they did in the first situation. The relationships between emotional social support seeking and instrumental social support seeking were not consistent, so these norms and behaviors were examined separately. Closer examination of the situational factor analyses provides some insight as to why three situations showed instrumental and emotional support seeking loading on different factors, whereas on two other situations, they loaded onto a single factor. One potential reason for the overlap is that both situations exhibited high levels of distress. Distress may evoke social support seeking behaviors from individuals that are both emotional and instrumental in nature. Examination of the situational content provides another reason: the two situations with overlap involve tension with coworkers. In situation 2, a coworker misinforms the participant about a meeting, which causes him/her to be late. In situation 3, the participant made a mistake that requires extra work from others. In these contexts, complaining about the problem to others in the workplace may be perceived negatively. Social support givers might be emotionally sympathetic only when the individual wants both emotional support and instrumental advice that could help mitigate the tension between colleagues. Although emotional and instrumental social support seeking behaviors were highly related to each other across situations, these relationships were lower than reported in past research. 48 A post-hoc examination of the mean differences in coping styles by situation showed each type of coping behaviors varied significantly across situations (emotional expression, Wilks’ Λ= .90, F (4, 127) = 3.55, p = .01; emotional social support seeking, Wilks’ Λ= .48, F (4, 126) = 33.84, p < .01; instrumental social support seeking, Wilks’ Λ= .59, F (4, 126) = 21.53, p <.01; use of humor, Wilks’ Λ= .73, F (4, 127) = 11.99, p < .01). _There is some literature that suggests different negative emotions have different origins and different underpinnings. For example, sadness is linked with situational attributions of negative events, whereas anger relates to a focus on the behaviors and intentions of individuals involved in a negative event (Keltner, Ellsworth, & Edwards, 1993). It may be that different types of negative emotions elicit different coping behaviors. Future studies should further explore how different emotions influence EFC behaviors and their relationships with various outcomes. The Relationship between Use of Humor Coping and Emotional Expression Use of Humor had inconsistent relationships with emotional expression behaviors across situations. It may be that the humor is not a universally appropriate reaction to negative events at work (for example, making jokes when your mistakes cause extra work for others), but emotional expression is more open-ended, and thus, more easily conceived of as tactful (for example, talking to coworkers about guilty feelings). The target, intent, and anticipation of the social reception of humor must be considered when assessing coping through the use of humor in future research. Deep Acting: Unexpected Linkages Deep Acting had several unexpected relationships with other variables in the study. It showed significant positive relationships with instrumental and emotional social support seeking, emotional expression, physical proximity, position tenure, and physical strain. Deep Acting 49 items focus on “really try[ing] to feel the emotions I have to show as part of my job” and “mak[ing] an effort” to show required emotions. The emphasis on feeling and showing required emotions implies some commitment to the job and the work involved. Individuals who work close with others and have been in their position for many years may feel more invested in displaying the emotions of interest. Individuals who share a commitment to their work may also be more likely to express emotions and seek support from others. Past research has examined commitment to display rules as having a moderating influence on the relationships between display rules and deep acting (Grosserand & Diefendorff, 2005), but more global forms of commitment (e.g. organizational or workgroup) commitment were not examined. Future research should examine what role organizational commitment plays in how individuals deal with their negative emotions, and whether commitment can explain some of the relationships found here. EFC Norms and Typical EFC Behavior Connections The connection between perceptions of emotion-focused coping norms and typical emotion-focused coping behaviors supported one of the study’s core hypotheses. These findings fit with past research indicating that perceptions of norms tend to predict norm conformity (Prentice & Miller, 1993). However, many participants reported a lack of insight into the emotional lives of other workgroup members, presenting a major concern. Several participants stated in the comments section that they had did not know how their coworkers deal with negative emotions, and openly acknowledged they wrote about their own emotion-focused coping when describing norms. A lack of insight into the emotion lives of others may have inflated the relationship between emotion-focused coping norms and typical emotion-focused coping behaviors. Some examples include: 50  “The questions regarding coworkers and colleagues emotions were difficult to answer since they are based on someone else actions/experiences.”  “The ‘how would your colleagues express their emotions questions’ were weird and difficult to answer... I just answered them how I felt.”  “It was interesting. I have never thought of co-workers emotions like that.”  “Hard to answer the questions about the emotions displayed by co-workers.”  “I found it difficult in the middle section, when trying to rate how I believed my staff would react when feeling emotions of guilt, distress, etc... in consulting, we move all over constantly, and many of the emotional issues that arise at work, happen when we're on the road or away from colleague, so it is tough sometimes to see how my staff react.” These comments indicate participants felt they did not know enough about their coworker’s emotion-focused coping to respond to the items with accuracy. Research has shown people tend to use information about themselves to describe others when information is low (Ready, Clark, Watson, & Westerhouse, 2000). Thus, emotion-focused coping norms’ significant predictions of typical coping behavior may have been somewhat artificially inflated by individuals’ projecting their own behaviors on how they think others behave. Future research could potentially address this by asking workers to rate the EFC behaviors of several specific coworkers with whom they work. This may make the measurement concrete, allowing participants to answer questions with more specificity. Nevertheless, many of the comments reflected a genuine appreciation and curiosity about the emotional cultures that differ across workplaces. Example comments are: 51  “I found this survey quite interesting because when I started my present job, I had never worked in such an environment and came from one in which emotions were usually expressed frequently and it was quite a learning adjustment for me to learn to control my emotions and reactions.”  “I would be interested to see your conclusions from peers in similar industries / departments. Emotion is taboo in most circles; colleagues are careful, political, and sometimes lack general human qualities. Too serious.”  “Kind of short, interesting to think of how people control things such as emotions, I work in a place that isn't too controlling but you still need to be careful what you say and not to offend anyone.” Based on these comments, it appears that many participants were aware of the existence of emotional cultures within their workplace, suggesting EFC norms likely do exist. One direction for future research comes from the climate strength literature. Climate strength research indicates that the influence of perceptions of a climate on employee outcomes may be moderated by the within-group variability of climate perceptions, such that groups with less climate variability have stronger climate-outcome relationships (Schneider, Salvaggio, & Subirats, 2002). Instead of relying solely on self-reported norms and individual behaviors, future research may be able to examine the group-level variability in how individuals perceive emotions being expressed and managed in the workplace. Beliefs about whether different emotion-related behaviors are appropriate and normal in the workplace may have a moderated influence on individuals’ behavioral outcomes, such that groups with low within-group variability have more dramatic effects. Groups with high variability in these beliefs (perhaps, 52 groups with members that serve very different job functions) may show weaker belief-outcome relationships. Although past research has looked the influence of unit-level display rule beliefs (Diefendorff et al., 2011), this research examined the main effect of display rules on both emotional regulation (surface/deep acting) and outcomes (employee well-being), and did not examine the moderating effect of variance in beliefs. Thus, research drawing from climate strength research would advance our understanding of this topic. Emotion-focused Coping Behaviors and Distal Employee Outcomes Emotion-focused coping behaviors did not significantly predict distal employee outcomes. This prevented the possibility of testing emotion-focused coping behavior as a mediator of norm-outcome relationships. This is surprising, given past research indicating positive effects of humor and emotional expression (respectively, Romero & Cruthirds, 2006; Pennebaker & Chung, 2011). Participant comments again provided several insights into why typical coping behavior over time might not link to distal outcomes. First, many participants noted that they do not frequently experience the negative events used in the typical coping measure. If individuals rarely experience negative events at work that elicit emotion-focused coping, it is unlikely coping strategies will influence the outcomes of organizational significance. Second, much of the research on the positive effects of emotional expression use samples of individuals coping with major illnesses (e.g. Stanton & Low, 2012). These situations are far more challenging and permanent than the typical negative events at work (for instance, getting a bad performance review). Third, participants commented that the coping items elicited a “yes, but…” response. This suggests the present study’s measure of typical emotion-focused coping was vague and did not capture sufficient information to realistically capture situational 53 responses. For example, emotional expression items ask individuals the extent to which would “let their feelings out.” But the item does not specify how public this expression was, what individuals were part of this expression, and what the repercussion of the expression was perceived to be. Subtleties like these may provide important indication of how and when emotion-focused coping strategies are used. Qualitative research might be able to better capture when emotion-focused coping occurs in the workplace, and how this subtly accrues over time. Trait Impulsivity and Self-Esteem as Moderators of Norm-Behavior Relationships The present study did not find any moderating effect of self-esteem on the relationship between EFC norms and EFC behaviors. Past research has shown the plasticity hypothesis is not always supported (Hui & Lee, 2000). If the relationship between EFC norms and EFC behaviors was inflated by lack of insight into the emotion-focused coping of others in general, this may be interfering with the present study’s ability to detect the effect of self-esteem on this relationship. Future research with group level data and more detailed measures could assess self-esteem’s true effect on this relationship. Most of the analyses testing impulsivity’ moderation showed null interaction terms. This may be due to the fact that impulsivity had an exceptionally low variance. One potential reason for this low variance in impulsivity is that the sample consisted of individuals who have secured employment, typically full-time, and many of whom are pursing higher education (e.g., the management course sample). Individuals who are highly impulsive tend to engage in socially deviant behaviors (Pratt & Cullen, 2000), and thus, would be less likely to have full time employment or to pursue higher education. Thus, the sample may not capture the full range of impulsivity in the general population. Trait impulsivity did in fact moderate the relationship between emotional social support seeking norms and typical emotional social support seeking 54 behavior in the hypothesized direction. This shows individuals who are not impulsive participate in emotional expression when others typically also engage in this type of coping, whereas highly impulsive individuals are less attuned to the environment. Future research should examine if impulsivity’s moderation holds when norms are assessed at the group level of analysis. Gender Gender did not moderate the relationship between typical coping through emotional expression and the positive outcomes as was expected. This prediction stemmed from Stanton et al.’s (2000) finding that emotionally expressive coping linked to a number of positive outcomes among women (e.g. lowered depressive symptoms), but not among men. However, gender did have significant relationships with many of the variables in the study. The situational data showed that across all situations, women were significantly more likely than men to report engaging in emotional social support seeking in response to the negative events. Women were also significantly more likely to engage in emotional expression coping in response to situation 1 (“colleague takes credit for your work”- high anger), whereas men were significantly more likely to engage in surface acting. Past research has suggested that women tend to complain in more expressive and indirect ways than men (Kowalski, 1996). It may be that men do not as readily engage in emotional expression as women do when they have been wronged by a coworker. Women were significantly more likely than men to report the presence of instrumental and emotional social support seeking, and emotional expression norms. Men were significantly more likely to perceive norms for surface acting. This connects to past propositions that the behaviors men view as complaining are often viewed by women as confiding (Kowalski, 1996). Women also reported experiencing more physical strain than men, a finding that matches past research (van Wijk & Kolk, 1997). 55 Limitations and Future Directions The present study had several limitations. First, norms were only measured at the level of perception. Future research could assess whether these norms are shared at higher levels of analysis (e.g., in workgroups, teams, or organizations). Second, common method variance is a concern with any survey design (Podsakoff et al., 2003). This was partially mitigated by the two survey format, but remains a concern. Third, one potential explanation for why this study’s results were not as expected is that the situations included are not encountered with great frequency in the workplace. In the present study, there was no way of knowing the frequency with which individuals encountered the selected situations or other negative events at work. This information would have provided a better understanding of the present study’s findings. Fourth, although Mechanical Turk is now viewed as a legitimate sample source for psychological research (Buhrmester et al., 2011), it requires researchers to trust Amazon’s workers to provide honest, accurate information. Researchers using Mechanical Turk do not have complete control over the context and sources of data collection. For example, several participants engaged in the primary study that was marked “full time employed participants only,” but marked within the study that they were actually unemployed or employed part time. These concerns were mitigated through strict data cleaning procedures (e.g., eliminating those who failed attention checks). The present study leads to several future research questions:  How frequent are negative events in non-service jobs in comparison to service jobs? One remaining question from this study is, how frequently are negative events encountered in this mostly non-service sample, and does this event frequency explain the study’s results? Event sampling methods could address this question. 56  When and why do non-service employees perform surface and deep acting? Study participants reported engaging in surface and deep acting behaviors in response to nonservice encounters. Emotional labor situations in non-service jobs may include interacting with higher-level employees or highly competitive coworkers.  Who are the key individuals in EFC strategies? Social network analysis could look at the influence of higher level employees or emotional confidants on the EFC behaviors of others.  How do perceptions of psychological safety and workgroup trust influence EFC norms and behaviors in the workplace? The linkages between emotional social support seeking, emotional expression, and surface acting behavior and norms show that emotional openness may vary across individuals and workplaces.  Is deep acting a reflection of organizational commitment? Deep acting was related to social support seeking behaviors, emotional expression, and position tenure. Organizational commitment could potentially account for these relationships, as employees who are familiar with others and are highly engaged in their work may make stronger efforts to meet perceived emotional obligations in the workplace.  Can EFC norms be assessed with greater granularity? The present study’s measure was highly specific in describing the behavioral antecedent, yet the EFC, surface acting, and deep acting behavior items lack specificity. Future measurement of EFC could assess how public coping is, what the perceived repercussions of emotional displays are in the situational context, and who provides social support (close friend, mentor, etc.)  Do the relationships between emotional social support seeking, emotional expression, and surface acting replicate beyond the present study? A review of the emotional labor 57 literature showed EFC behaviors, surface acting, and deep acting behaviors have not been studied in simultaneously in the past. Additional research could explore whether these relationships replicate beyond the current study.  When do social support-emotional and social support-instrumental coincide at work? The present study showed these behaviors were related, but not as highly as in past research. Future studies could examine whether instrumental and emotional social support seeking behaviors coincide when participants respond to a larger array of negative events.  Do the variance in perceptions of EFC norms and emotional beliefs strengthen the linkages between EFC norm perceptions and outcomes? Drawing from the climate strength literature, within-group variance may influence whether or not norm perceptions have influence on EFC behaviors and other outcomes.  Do shared norms have effects on individuals that are moderated by individual self-esteem or trait impulsivity? The present study’s findings did not support hypotheses regarding self-esteem and impulsivity’s moderation of norm-behavior relationships. However, future research could examine whether this is the case when norms are aggregated within workgroups. Concluding Remarks In the present study, EFC norms showed connections to EFC behaviors, but EFC behaviors did not exhibit linkages to distal employee outcomes. The present research supports the notion that injunctive display rules and descriptive EFC norms are distinct constructs. It also identified linkages between emotional expression, emotional social support seeking, and surface acting, suggest constructs from the coping literature may overlap with emotional labor constructs. Self-esteem did not moderate norm-behavior relationships; trait impulsivity 58 moderated the relationships between emotional social support seeking norms and typical emotional social support seeking behaviors. Counter to what was expected, gender did not moderate emotional expression behaviors’ influence on distal outcomes; however, gender was related to a variety of coping behaviors. 59 APPENDICES 60 Appendix A: Pilot study consent form Negative Emotions in the Workplace In this research questionnaire, we will be asking you to respond to a series of questions about emotions experienced at work. We expect that it will take about 10 minutes for you to complete this survey. You will receive 1 HPR credit as compensation upon completion of this survey. There are no foreseeable risks associated with participating in this study. The survey will be entirely anonymous—that is, we will not be able to connect your responses to any identifying information. Your confidentiality will be protected to the maximum extent allowable by law. The data will be saved for at least five years after it is collected and will only be accessible by the primary investigator. Participation is completely voluntary. You may refuse to participate in certain procedures or answer certain questions. You may choose not to participate at all and may discontinue your participation at any time without penalty or loss of benefits. Other HPR experiments and alternative options for course credit are available to you. Please contact your instructor for specific details. If you have any questions or concerns about your participation in this project, you can reach Catherine Ott-Holland by phone: (262) 497-2486, email: ottholla@msu.edu, or regular mail: 348 Psychology Building, East Lansing, MI 48824. If you have any questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this research study, you may contact, anonymously if you wish, the Michigan State University Human Research Protection Program at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu, or regular mail at: 408 W. Circle Dr., 207 Olds Hall, MSU, East Lansing, MI 48824. Please mark the box that says "I agree to give my consent to participate" if you agree to participate in this study. If you do not agree to participate in this study, you may exit this website now. Thank you, Catherine Ott-Holland  I agree to give my consent to participate. 61 Appendix B: Pilot study ratings People experience a range of different emotions in the workplace. With each of the following situations, think of a time when you might have experienced this type of event at work. If you have not experienced an event on this list while at work, think of how you might feel if the event were to occur. Rate the extent the event has or would have elicited the following emotions. Situation 1: You discover a colleague has taken credit for your work without permission. Situation 2: A supervisor provides you with harsh feedback in front of a group of your peers. Situation 3: A device necessary to complete an important assignment breaks unexpectedly. Situation 4: A colleague gives you incorrect information about a meeting, causing you to show up an hour late. Situation 5: Someone in your department is fired, resulting in additional work for everyone remaining. Situation 6: Your boss gives you additional work when you are already over-committed with tasks. Situation 7: A reward you were expecting (promotion, raise, bonus, etc.) was not given to you as promised by the organization. Situation 8: You learn a mistake you made at work resulted in additional work for several other employees. Situation 9: Something on which you were working was discarded by the organization Situations 10: A coworker makes mean-spirited comments about your recent work. Situation 11: You hear rumors that there will be a layoff in your department. 62 Situation 12: A new boss is taking over your department that is known to be tough and demanding. Situation 13: You learn your department did not win a company-wide competition. Ratings: Anger Sadness Fear Distress Guilt 1= Not at all 2= Slightly 3= Moderately 4=Very 5=Extremely 63 Appendix C: Primary study survey 1 online consent form--management course version Consent form and instructions This is first questionnaire for a two-part study on negative emotions in the workplace. You must be 18 or older to participate in this study. You will receive 1 points extra credit if you participate in this survey, and 2 extra credit points if you participate in the second survey, making a total of 3 possible extra credit points. The second survey will be sent out one week after the initial survey is sent. It is very important to our research that you respond to both questionnaires. We will be asking you to respond to a series of questions about negative emotions you and others have felt while at work, and about how you feel about your job and organization. We are also asking you to respond to some commonly used demographic questions that will help us interpret the meaning of your responses to the questionnaire. We expect that it will take about 20 minutes for you to complete this survey. There are no foreseeable risks associated with participating in this study. You will not be asked for any identifying information other than a unique ID number. Your privacy will be protected to the maximum extent allowable by law. The data will be saved for at least five years after it is collected and will only be accessible by the primary investigator and one graduate student. By marking below, you indicate that you are free to refuse to participate in this project or any part of this project. You may refuse to participate in certain procedures or answer certain questions. Participation is completely voluntary. You may choose not to participate at all and may discontinue your participation at any time without penalty or loss of benefits. If you have any questions or concerns about your participation in this project, you can reach Catherine Ott-Holland by phone: (262) 497-2486, email: ottholla@msu.edu, or regular mail: 348 Psychology Building, East Lansing, MI 48824. Please mark the box that says "I agree to give my consent to participate" if you agree to participate in this study. □ I agree to give my consent to participate. 64 Appendix D: Primary study survey 1 online consent form—Mechanical Turk version Consent form and instructions This is first questionnaire for a two-part study on negative emotions in the workplace. You must be 18 or older to participate in this study. You will receive $2.00 for participating in part 1. If you complete survey 1 fully, you will be invited to participate in a second survey within the next week. The compensation for the second survey will be $3.00. It is very important to our research that you respond to both questionnaires. We will be asking you to respond to a series of questions about negative emotions you and others have felt while at work, and about how you feel about your job and organization. We are also asking you to respond to some commonly used demographic questions that will help us interpret the meaning of your responses to the questionnaire. We expect that it will take about 20 minutes for you to complete this survey. There are no foreseeable risks associated with participating in this study. You will not be asked for any identifying information other than a unique ID number. Your privacy will be protected to the maximum extent allowable by law. The data will be saved for at least five years after it is collected and will only be accessible by the primary investigator and one graduate student. By marking below, you indicate that you are free to refuse to participate in this project or any part of this project. You may refuse to participate in certain procedures or answer certain questions. Participation is completely voluntary. You may choose not to participate at all and may discontinue your participation at any time without penalty or loss of benefits. If you have any questions or concerns about your participation in this project, you can reach Catherine Ott-Holland by phone: (262) 497-2486, email: ottholla@msu.edu, or regular mail: 348 Psychology Building, East Lansing, MI 48824. Please mark the box that says "I agree to give my consent to participate" if you agree to participate in this study. □ I agree to give my consent to participate. 65 Appendix E: Primary study survey 2 consent form—management course version This is second questionnaire for a two-part study on negative emotions in the workplace. You must be 18 or older to participate in this study. We will be asking you to respond to a series of questions about negative emotions you and others have felt while at work, and about how you feel about your job and organization. We expect that it will take about 10 minutes for you to complete this survey. You will receive extra credit for completing this survey. There are no foreseeable risks associated with participating in this study. You will receive 2 extra credit points for your participation. Your privacy will be protected to the maximum extent allowable by law. The data will be saved for at least five years after it is collected and will only be accessible by the primary investigator and one graduate student. By marking below, you indicate that you are free to refuse to participate in this project or any part of this project. You may refuse to participate in certain procedures or answer certain questions. Participation is completely voluntary. You may choose not to participate at all and may discontinue your participation at any time without penalty or loss of benefits. If you have any questions or concerns about your participation in this project, you can reach Catherine Ott-Holland by phone: (262) 497-2486, email: ottholla@msu.edu, or regular mail: 348 Psychology Building, East Lansing, MI 48824. Please mark the box that says "I agree to give my consent to participate" if you agree to participate in this study. □ I agree to give my consent to participate. 66 Appendix F: Primary study survey 2 consent form—Mechanical Turk version This is second questionnaire for a two-part study on negative emotions in the workplace. You must be 18 or older to participate in this study. We will be asking you to respond to a series of questions about negative emotions you and others have felt while at work, and about how you feel about your job and organization. We expect that it will take about 10 minutes for you to complete this survey. You will be compensated $3.00 for completing this survey as a bonus through Mechanical Turk. There are no foreseeable risks associated with participating in this study. Your privacy will be protected to the maximum extent allowable by law. The data will be saved for at least five years after it is collected and will only be accessible by the primary investigator and one graduate student. By marking below, you indicate that you are free to refuse to participate in this project or any part of this project. You may refuse to participate in certain procedures or answer certain questions. Participation is completely voluntary. You may choose not to participate at all and may discontinue your participation at any time without penalty or loss of benefits. If you have any questions or concerns about your participation in this project, you can reach Catherine Ott-Holland by phone: (262) 497-2486, email: ottholla@msu.edu, or regular mail: 348 Psychology Building, East Lansing, MI 48824. Please mark the box that says "I agree to give my consent to participate" if you agree to participate in this study, then click "continue." □ I agree to give my consent to participate. 67 Appendix G: Self Esteem Rate the extent to which you believe the following statements are true for you. 1= Not at all True of Me 2= Slightly True of Me 3= Somewhat True of Me 4= True of Me to a Great Extent 1. I feel that I am a person of worth, at least on an equal plane with others. 2. I feel that I have a number of good qualities. 3. All in all, I am inclined to feel that I am a failure. 4. I am able to do things as well as most other people. 5. I feel I do not have much to be proud of. 6. I take a positive attitude toward myself. 7. On the whole, I am satisfied with myself. 8. I wish I could have more respect for myself. 9. I certainly feel useless at times. 10. At times I think I am no good at all. 68 Appendix H: Impulsivity DIRECTIONS: People differ in the ways they act and think in different situations. This is a test to measure some of the ways in which you act and think. Read each statement and mark an X on the appropriate circle on the right side. Do not spend too much time on any statement. Answer quickly and honestly. 1= Rarely/Never 2= Occasionally 3= Often 4= Almost Always 1. I plan tasks carefully. 2. I do things without thinking. 3. I make-up my mind quickly. 4. I am happy-go-lucky. 5. I don’t “pay attention.” 6. I have “racing” thoughts. 7. I plan trips well ahead of time. 8. I am self-controlled. 9. I concentrate easily. 10. I save regularly. 11. I “squirm” at plays or lectures. 12. I am a careful thinker. 13. I plan for job security. 14. I say things without thinking. 15. I like to think about complex problems. 16. I change jobs. 17. I act “on impulse.” 18. I get easily bored when solving thought problems. 19. I act on the spur of the moment. 20. I am a steady thinker. 21. I change residences. 22. I buy things on impulse. 23. I can only think about one thing at a time. 24. I change hobbies. 25. I spend or charge more than I earn. 26. I often have extraneous thoughts when thinking. 27. I am more interested in the present than the future. 28. I am restless at the theater or lectures. 29. I like puzzles. 30. I am future oriented. 69 Appendix I: Workgroup Coping Norms Imagine how the people you work with generally deal with negative feelings, such as sadness, anger, fear, distress, or guilt. Rate the following items based on the extent to which you think individuals in your workgroup would use them to respond to negative feelings. 1= Strongly Disagree 2= Disagree 3= Neutral 4= Agree 5= Strongly Agree When the people in your department that you work with regularly experience sadness, anger, or fear, they generally… (Seeking social support) 1. Ask people who have had similar experiences what they did. 2. Try to get advice from someone about what to do. 3. Talk to someone to find out more about the situation. 4. Talk to someone who could do something concrete about the problem. 5. Talk to someone about how they feel. 6. Try to get emotional support from friends or relatives. 7. Discuss their feelings with someone. 8. Get sympathy and understanding from someone. (Using humor) 9. Find it hard to laugh it off with other people. 10. Make jokes to lighten the mood. 11. Laugh it off until they feel better. 12. Use humor to cope with their negative feelings. (Emotional Expression)) 13. Let their feelings come out freely. 14. Take time to express their emotions. 15. Allow themselves to express their emotions. 16. Feel free to express their emotions. When people in my workgroup experience sadness, anger, or fear, they generally… (Suppression Items --for comparison) 17. Keep their emotions to themselves 18. Control their emotions by not expressing them 19. Make sure not to express the negative emotions they are feeling. (Surface and Deep Acting Items- for comparison) 20. Resist expressing their true feelings 21. Pretend to have emotions that they don’t really have 70 22. Hide their true feelings about a situation 23. Make an effort to actually feel the emotions they need to display to others 24. Try to actually experience the emotions that they must show 25. Really try to feel the emotions they have to show as part of their job. 71 Appendix J: Display Rules (Requirement to Hide Negative Emotions) Among the individuals I regularly interact with in my department, rate the extent to which you are required to hide emotion in order to be effective on the job. 1= Not at all required 2= Rarely Required 3= Sometimes Required 4= Usually Required 5= Always Required 1. Hide your anger or disapproval about something someone has done (e.g. an act that is distasteful to you.) 2. Hide your disgust over things others have done. 3. Hide your fear of someone who appears threatening. 72 Appendix K: Demographics Please answer the following questions about yourself: I am (circle one): MALE FEMALE Work Experience: _____ years, _____ months How long have you been in your current position? ____years, ____ months Industry: ____________ My age is: ___________ My ethnicity is (check one or more): American Indian or Alaska Native Asian Black or African American Hispanic or Latino Native Hawaiian or Pacific Islander Caucasian Other: ______________________ 73 Appendix L: Coping Behaviors The following situations represent events that typically lead people to feel negative emotions at work. Different people deal with these negative feelings at work in different ways. Think of how you generally react to the each of these situations when they occur in your workplace. If this situation has never happened to you, please mark the box stating that the event has never previously happened to you at work. Then, imagine how you would likely react to the situation and rate the items that follow accordingly. Situation 1 You discover a colleague has taken credit for your work without permission. Situation 2 A colleague gives you incorrect information about a meeting, causing you to show up an hour late. Situation 3 A reward you were expecting (promotion, raise, bonus, etc.) was not given to you as promised by the organization. Situation 4 You learn a mistake you made at work resulted in additional work for several other employees. Situation 5 You hear rumors that there will be a layoff in your department.  Check here if this situation has never happened to you at work. In general I would feel… Anger Sadness Fear Distress Guilt 1= Not at all 2= Slightly 3= Moderately 4=Very 5=Extremely 74 In general I would… 1= Never 2= Rarely 3= Every Once in a While 4= Sometimes 5= Almost always (Seeking social support) 1. Ask people who have had similar experiences what they did. 2. Try to get advice from someone about what to do. 3. Talk to someone to find out more about the situation. 4. Talk to someone who could do something concrete about the problem. 5. Talk to someone about how I feel. 6. Try to get emotional support from friends or relatives. 7. Discuss my feelings with someone. 8. Get sympathy and understanding from someone. (Using humor) 9. Find it hard to laugh about the situation with other people. 10. Make jokes to lighten the mood. 11. Laugh it off until I feel better. 12. Use humor to cope with my negative feelings. (Emotional Expression) 13. Let my feelings come out freely. 14. Take time to express my emotions. 15. Allow myself to express my emotions. 16. Feel free to express my emotions. In this situation, in general I would… (Suppression Items --for comparison) 17. Keep my emotions to myself 18. Control my emotions by not expressing them 19. Make sure not to express the negative emotions I am feeling. (Surface and Deep Acting Items- for comparison) 20. Resist expressing my true feelings 21. Pretend to have emotions that I don’t really have 22. Hide my true feelings about a situation 23. Make an effort to actually feel the emotions I need to display to others 24. Try to actually experience the emotions that I must show 25. Really try to feel the emotions I have to show as part of my job 75 Appendix M: Physical Strain Symptoms The following questions ask about how you feel physically over the past three months. Please use the scale provided to indicate how often you have experienced each symptom by choosing the frequency indicators, and checking the circle that best represent your response. 1= Less than Once per Month or Never 2= Once or Twice per Month 3= Once or Twice per Week 4= Once or Twice per Day 5= Several Times per Day 1. An upset stomach or nausea 2. A backache 3. Trouble sleeping 4. Headache 5. Acid indigestion or heartburn 6. Eye strain 7. Diarrhea 8. Stomach cramps (Not menstrual) 9. Constipation 10. Ringing in the ears 11. Loss of appetite 12. Dizziness 13. Tiredness or fatigue 76 Appendix N: Workgroup Helping Intentions Answer the following based on how you would be willing to engage in the following actions for the people in your department that you work with regularly. 1= Very Probably Not 2= Probably Not 3= Possibly 4= Probably 5= Very Probably 6= Definitely I would be willing to… Volunteer to do things for my workgroup. Attend a function that helps my workgroup. Assist others in the group with their work for the benefit of the group. Get involved to benefit my work group. Help others in my workgroup learn about the work. Help others in my workgroup with their work responsibilities. 77 Appendix O: Satisfaction with Coworkers Rate the extent to which the following adjectives describe the people with whom you work in general. 1= Strongly Disagree 2= Disagree 3= Neutral 4= Agree 5= Strongly Agree People on Your Present Job Stimulating Boring Slow Ambitious Stupid Responsible Fast Intelligence Easy to make enemies Talk too much Smart Lazy Unpleasant No privacy Active Narrow interests Loyal Hard to meet 78 Appendix P: Workgroup P-E Fit Rate the extent to which you agree with the following statements about the people in your department that you work with regularly. 1= Disagree 2= Undecided 3= Agree 1. The things that I value in life are very similar to the things that my workgroup members value. 2. My personal values match my workgroup’s values and culture. 3. My workgroup’s values and culture provide a good fit with the things that I value in life. 4. I feel my personality matches my workgroup’s image. 5. My personality matches my workgroup’s personality. 6. My personality provides a good fit with my workgroup’s personality. 79 Appendix Q: Tables and Figures Table 1 Definitions of Coping Source Pearlin & Schooler (1978) Coping Definition “Any response to external life strains that serves to prevent, avoid or control emotional distress.” Folkman & Lazarus (1980) “Coping is defined as the cognitive and behavioral efforts made to master, tolerate or reduce external and internal demands and conflicts among them.” “A conscious, intentional, goal directed response, tailored to the specific demands of a stressor” Lazarus & Folkman (1984) Connor-Smith & Flachsbart (2007) “Conscious, volitional attempts to regulate the environment or one’s reaction to the environment under stressful conditions.” 80 Table 2 Emotion-focused Coping Dimensions Dimension Source Definition Seeking social support for instrumental reasons COPE Measure (Carver et al., 1989) Seeking advice, assistance, or information. Seeking social support for emotional reasons COPE Measure (Carver et al., 1989) Getting moral support, sympathy, or understanding. Use of humor Martin and Leftcourt (1983) Degree to which subjects report using humor as a means of coping with stressful experiences. Emotional Expression Stanton et al. (2000) Interpersonal and intrapersonal (e.g. journal writing, artistic production) forms of expression 81 Table 3 Ratings of the emotional reactions to negative workplace situations Anger M SD Workplace Situation 1) You discover a colleague has taken credit for your work without permission.* 2) A supervisor provides you with harsh feedback in front of a group of your peers. 3) A device necessary to complete an important assignment breaks unexpectedly. 4) A colleague gives you incorrect information about a meeting, causing you to show up an hour late.* 5) Someone in your department is fired, resulting in additional work for everyone remaining. 6) Your boss gives you additional work when you are already over-committed with tasks. 7) A reward you were expecting (promotion, raise, bonus, etc.) was not given to you as promised by the organization.* 8) You learn a mistake you made at work resulted in additional work for several other employees.* 9) Something on which you were working was discarded by the organization 10) A coworker makes mean-spirited comments about your recent work. 11) You hear rumors that there will be a layoff in your department.* 12) A new boss is taking over your department that is known to be tough and demanding. 82 Sadness M SD Fear M SD Distress M SD Guilt M SD 3.71 0.87 2.02 0.81 1.29 0.59 2.51 1.18 1.24 0.53 3.11 1.11 2.84 1.07 2.24 1.03 2.98 1.12 2.09 0.9 3.32 1.18 2.20 1.16 2.41 1.21 3.27 1.25 0.84 3.89 0.89 1.84 0.9 1.6 2.49 1.16 3.04 1.22 1.82 1.11 2.27 1.03 2.13 1.01 1.69 0.85 2.47 1.12 1.51 0.87 3.02 0.95 1.73 0.87 1.93 0.97 3.34 1.12 1.3 0.63 3.96 1.07 3.13 1.18 1.56 0.84 2.73 1.32 1.42 0.92 2.24 1.21 2.93 1.18 2.64 3.29 1.1 1.1 3.2 1.08 3.87 0.97 3.13 0.99 1.82 1.09 2.7 1.19 1.62 0.94 3.47 1.08 2.69 2.42 1.1 1.1 1.56 0.92 2.58 1.1 1.53 0.79 2.93 1.25 3.78 1.02 3.47 1.1 1.62 0.86 1.82 1.03 2.18 1.17 2.93 1.1 2.71 1.12 1.41 0.73 Table 3 (cont’d) 13) You learn your department did not win a company-wide 1.44 0.73 2.09 0.9 1.07 0.25 1.49 0.7 1.38 0.54 competition. Note. Average emotion ratings of 3 or above are marked in bold above. Situations that were selected for the primary data collection are marked with an asterisk. 83 Table 4 Descriptive Information for Demographic Variables Demographic Variables N Minimum Maximum Mean SD Gender 148 0 1 0.62 0.49 Race 148 0 1 0.78 0.41 Work Experience 148 3 40 16.36 9.22 Level of Employment 149 0 1 0.97 0.18 Proximity to Colleagues 149 1 3 2.64 0.68 Tenure in current position 149 0 23 4.77 4.00 Note. Gender is coded 0=women, 1=men; Race is coded 0=Not-White, 1= White; Level of Employment is coded 0=3/4 time employment, 1=full time (approximately 40 hours per week); Proximity to Colleagues is coded 1= I work virtually outside of the office, 2= I work in an office, but not in close proximity to my colleagues, and 3= I work in an office in close physical proximity to my colleagues; Work experience and Tenure in current position were measured in years. 84 Table 5 Varimax Rotated Maximum Likelihood Factor Structure for All 25 Norm Items 1 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 Emotional Social Support 1 Emotional Social Support 2 Emotional Social Support 3 Emotional Social Support 4 Use of Humor 1 Use of Humor 2 Use of Humor 3 Use of Humor 4 Emotional Expression 1 Emotional Expression 2 Emotional Expression 3 Emotional Expression 4 Emotional Suppression 1 Emotional Suppression 2 Emotional Suppression 3 Surface Acting 1 Surface Acting 2 Surface Acting 3 Deep Acting 1 Deep Acting 2 Deep Acting 3 Display Rules 1 Display Rules 2 Display Rules 3 2 Factor 3 4 0.52 0.75 0.86 0.79 5 0.73 0.85 0.73 0.35 6 7 0.35 0.47 0.84 0.88 0.82 -0.34 -0.31 -0.41 -0.42 0.84 0.91 0.71 0.80 0.57 0.58 0.80 0.74 0.69 0.77 0.86 0.72 14.17 Variance Explained % Note. N=143; Loadings under .30 were suppressed. 9.36 % 85 9.13 % 8.34 % 8.18 % 0.79 0.98 0.62 7.68 % 7.55 % Table 6 Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 1 1 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 Emotional Social Support 1 Emotional Social Support 2 Emotional Social Support 3 Emotional Social Support 4 Use of Humor 1 Use of Humor 2 Use of Humor 3 Use of Humor 4 Emotional Expression 1 Emotional Expression 2 Emotional Expression 3 Emotional Expression 4 Emotional Suppression 1 Emotional Suppression 2 Emotional Suppression 3 Surface Acting 1 Surface Acting 2 Surface Acting 3 Deep Acting 1 Deep Acting 2 Deep Acting 3 2 -0.39 0.58 0.85 0.84 0.74 Factor 3 4 0.86 0.91 0.57 0.54 0.31 5 6 0.47 0.88 0.89 0.93 -0.51 -0.47 -0.45 0.77 0.86 0.84 0.82 0.56 0.85 0.54 0.61 0.72 0.78 0.33 -0.31 0.79 0.93 0.82 19.08% 12.33% 10.83% 10.44% 9.04% 8.75% Variance Explained Note. N=132; Loadings under .30 were suppressed. Situation 1 was “You discover a colleague has taken credit for your work without permission.” 86 Table 7 Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 2 Factor 1 2 0.64 0.69 0.52 0.43 0.81 0.84 0.89 0.83 0.36 3 4 5 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 Emotional Social Support 1 0.32 Emotional Social Support 2 Emotional Social Support 3 Emotional Social Support 4 Use of Humor 1 0.37 Use of Humor 2 0.92 Use of Humor 3 0.88 Use of Humor 4 0.88 Emotional Expression 1 -0.37 0.52 Emotional Expression 2 -0.36 0.62 Emotional Expression 3 -0.44 0.32 0.77 Emotional Expression 4 -0.44 0.81 Emotional Suppression 1 0.86 Emotional Suppression 2 0.85 Emotional Suppression 3 0.84 Surface Acting 1 0.82 Surface Acting 2 0.71 Surface Acting 3 0.84 Deep Acting 1 0.80 Deep Acting 2 0.96 Deep Acting 3 0.87 19.18% 19.09% 10.44% 10.23% 9.78% Variance Explained Note. N=143; Loadings under .30 were suppressed. Situation 2 was “Your colleague gives you incorrect information about a meeting, causing you to show up an hour late.” 87 Table 8 Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 3 1 2 Factor 3 4 0.79 0.89 0.69 0.51 0.38 5 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 Emotional Social Support 1 0.74 Emotional Social Support 2 0.80 Emotional Social Support 3 0.84 Emotional Social Support 4 0.79 Use of Humor 1 0.31 0.38 Use of Humor 2 0.94 Use of Humor 3 0.97 Use of Humor 4 0.93 Emotional Expression 1 -0.60 0.30 0.31 Emotional Expression 2 -0.54 0.44 0.33 Emotional Expression 3 -0.66 0.44 Emotional Expression 4 -0.72 0.32 Emotional Suppression 1 0.95 Emotional Suppression 2 0.94 Emotional Suppression 3 0.90 Surface Acting 1 0.87 Surface Acting 2 0.64 Surface Acting 3 0.80 Deep Acting 1 0.95 Deep Acting 2 0.95 Deep Acting 3 0.86 Variance Explained 25.43% 14.15% 11.82% 11.53% 11.04% Note. N=138; Loadings under .30 were suppressed. Situation 3 was “A reward you were expecting (promotion, raise, bonus, etc.) was not given to you as promised by the organization.” 88 Table 9 Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 4 1 2 0.60 0.55 0.55 0.38 0.85 0.85 0.88 0.79 Factor 3 4 5 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 0.34 Emotional Social Support 1 Emotional Social Support 2 Emotional Social Support 3 Emotional Social Support 4 Use of Humor 1 Use of Humor 2 0.86 Use of Humor 3 0.95 Use of Humor 4 0.90 Emotional Expression 1 -0.45 0.38 0.66 Emotional Expression 2 -0.47 0.42 0.70 Emotional Expression 3 -0.45 0.38 0.75 Emotional Expression 4 -0.45 0.37 0.74 Emotional Suppression 1 0.87 -0.33 Emotional Suppression 2 0.91 Emotional Suppression 3 0.80 Surface Acting 1 0.88 Surface Acting 2 0.56 Surface Acting 3 0.81 Deep Acting 1 0.91 Deep Acting 2 0.95 Deep Acting 3 0.92 Variance Explained 20.24% 19.87% 11.35% 10.83% 10.30% Note. N=138; Loadings under .30 were suppressed. Situation 4 was “You learn a mistake you made at work resulted in additional work for several employees.” 89 Table 10 Varimax Rotated Maximum Likelihood Factor Structure for Coping in Situation 5 1 2 Factor 3 4 5 0.85 0.91 0.38 0.55 0.36 Instrumental Social Support 1 Instrumental Social Support 2 Instrumental Social Support 3 Instrumental Social Support 4 Emotional Social Support 1 0.63 Emotional Social Support 2 0.83 Emotional Social Support 3 0.83 Emotional Social Support 4 0.78 Use of Humor 1 0.37 0.40 Use of Humor 2 0.91 Use of Humor 3 0.94 Use of Humor 4 0.89 Emotional Expression 1 -0.57 0.47 Emotional Expression 2 -0.57 0.56 Emotional Expression 3 -0.53 0.56 Emotional Expression 4 -0.53 0.55 Emotional Suppression 1 0.89 Emotional Suppression 2 0.91 Emotional Suppression 3 0.86 Surface Acting 1 0.86 Surface Acting 2 0.61 Surface Acting 3 0.79 Deep Acting 1 0.90 Deep Acting 2 0.95 Deep Acting 3 0.93 32.38% 14.01% 10.27% 8.58% 6.14% Variance Explained Note. N=138; Loadings under .30 were suppressed. Situation 5 was “You hear rumors that there will be a layoff in your department.” 90 Table 11 Descriptive Statistics of Manipulation Checks by Situation Situation 1 Situation 2 Situation 3 Situation 4 Situation 5 Anger M 3.73 3.40 3.97 2.46 2.55 SD .90 1.25 1.02 1.12 1.16 Sadness M 2.13 1.77 3.21 3.11 3.17 SD 1.10 1.01 1.18 1.17 1.13 Fear M 1.26 2.17 1.73 2.68 3.82 SD .54 1.29 .99 1.19 1.14 Distress M 2.53 3.03 3.13 3.48 3.73 SD 1.22 1.29 1.25 1.09 1.08 Guilt M 1.08 1.75 1.31 4.09 1.53 SD .30 .99 .70 .88 .93 Note. Situation 1 was “You discover a colleague has taken credit for your work without permission.” Situation 2 was “A colleague gives you incorrect information about a meeting, causing you to show up an hour late.” Situation 3 was “A reward you were expecting (promotion, raise, bonus, etc. ) was not given to you as promised by the organization.” Situation 4 was “You learn a mistake you made at work resulted in additional work for several other employees.” Situation 5 was “You hear rumors that there will be a layoff in your department.” 91 Table 12 Situation-Specific Typical EFC, SA, and DA Behavior Descriptives, Reliabilities, and Norm Intercorrelations Typical Behaviors Situation 1 SSI SSE Humor EE SA DA Situation 2 SSI SSE Humor EE SA DA Situation 3 SSI SSE Humor EE SA DA M SD Within –Situation Behavior Norms SSI SSE Humor EE SA DA SSI SSE Humor EE SA DA DR 3.37 3.16 2.74 3.12 2.53 2.76 0.92 (.85) .46* 1.07 (.90) 0.99 0.87 0.98 1.02 -.01 -.02 (.93) .34* .61* -.03 (.91) -.05 -.27* .06 -.72* (.84) .13 .18* -.03 .13 .03 (.89) .18* -.02 .10 .10 -.16 .19* .05 .31* -.05 .31** -.28** .17* .19* -.14 .23* -.09 .04 .13 <.01 .24* .06 .44** -.43* .21* .05 -.14 -.05 -.30* .38** -.12 <.01 .09 <.01 .19* -.17* .10 .03 -.08 -.05 -.11 .14 .05 2.87 2.65 2.72 3.04 2.56 2.76 1.04 (.86) .58* 1.17 (.94) 1.10 0.90 1.02 1.02 .17 .13 (.81) .22* .48* .12 (.91) -.03 -.09 -.05 -.73* (.88) .32* .22* .03 -.01 .22* (.91) .18* .04 -.08 .07 .01 .19* .03 .27* -.16 .27* -.17* .17* .16 -.03 .15 -.11 .08 .13 -.11 .16 -.05 .28* -.26* .21* .07 -.13 -.05 .06 .08 .01 -.20* .17 .23* -.13 -.12 .10 .18* .07 .09 -.12 .11 .05 3.61 3.39 2.24 3.20 2.64 2.71 0.95 (.86) .53* 1.11 (.92) 1.02 0.96 1.05 1.14 .04 .12 (.97) .47* .51* .07 (.94) -.31* -.32* .03 -.80* (.88) .17* .17* .17* .15 .04 (.95) .21* .11 -.02 .36* .02 .05 .08 .32* .02 -.23* .06 .02 .17* -.08 .17* -.04 .10 .13 .14 .35* .06 .35* -.29* .11 -.15 -.24* .10 -.33* .34* -.06 <-.01 -.02 .08 -.14 .17* .06 92 <-.01 .08 .09 .19* -.13 .17* Table 12 (cont’d) Note. *p<.05. SSI=Social Support Seeking-Instrumental, SSE=Social Support Seeking-Emotional. EE=Emotional Expression, SA=Surface Acting, DA=Deep Acting, DR=Display Rules. Norms were measured using a 5-point scale of agreement. Typical EFC, surface and deep acting behaviors were measured using a 5-point frequency scale. Within-situation correlation reflect the relationships between typical behaviors, surface and deep acting dimensions within the situation listed in the row. Reliabilities are listed in parentheses. Correlations between related norms and behaviors are marked in bold. Situation 1 was “You discover a colleague has taken credit for your work without permission.” Situation 2 was “A colleague gives you incorrect information about a meeting, causing you to show up an hour late.” Situation 3 was “A reward you were expecting (promotion, raise, bonus, etc. ) was not given to you as promised by the organization.” Situation 4 was “You learn a mistake you made at work resulted in additional work for several other employees.” Situation 5 was “You hear rumors that there will be a layoff in your department.” 93 Table 12 (cont’d) Typical Behaviors Situation 4 SSI Situation 4 SSE Situation 4 Humor Situation 4 EE Situation 4 SA Situation 4 DA Situation 4 Situation 5 SSI Situation 5 SSE Situation 5 Humor Situation 5 EE Situation5 SA Situation5 DA Situation 5 Within –Situation Behavior SSE Humor EE SA DA SSI Norms SSE Humor EE -.39* -.24* -.14 -.75* (.87) .27* .28* .14 .19* .07 (.97) .04 -.12 -.04 -.15 .14 .09 -.02 .26* -.19* .10 -.05 .06 .11 .01 .11 -.04 .08 .17* -.05 .22* -.08 .21* -.16 .09 .05 -.06 .14 -.14 .21* -.08 .37* -.11 .56* -.22* .19* <0.01 (.94) -.71* (.86) .18* .11 .16 .17* .17* (.96) .24* -.02 .10 -.01 .08 .27* -.07 .29* .05 .24* -.20* .10 .06 -.14 .28* -.08 .07 .15 .02 .29* .05 .37* -.19* .15 -.06 .01 .15 -.14 .11 .04 .02 <-.01 .04 -.34* .22* -.18* .25* -.19* .32* -.12 .18* .21* M SD SSI 3.19 2.74 2.24 2.94 2.75 2.87 1.11 1.22 1.05 1.05 1.09 1.17 (.92) .53* (.93) .24* .11 (.77) .52* .51* .26* (.95) 3.76 3.37 2.38 3.07 2.63 2.85 0.87 1.10 1.05 0.98 1.04 1.17 (.81) .40* (.90) .08 .12 (.94) 94 SA DA DR -.01 .07 .07 .11 -.19* -.04 .18* .13 .22* .02 .14 .20* Table 13 Situation-Specific Typical EFC, SA, and DA Behavior Inter-correlations with Outcome and Remaining Variables Typical Phys. P-E WG Work Phys. HelpI SE Impuls. Gend Race Behavior Strain Fit Sat. Exp. Prox. Sit. 1 SSI <-.01 .04 .11 .10 .13 -.16 -.01 .10 -.12 .04 SSE .15 <.01 -.05 .04 -.06 -.01 -.31* .14 -.08 -.08 * Humor -.12 .05 .06 .13 .11 .11 .14 .19 -.08 -.03 EE .03 -.06 -.07 -.04 .07 -.06 -.20* .10 .08 -.04 SA .07 .05 .02 .02 -.14 .03 .20* -.09 -.17* .03 DA .13 .02 >-.01 -.03 -.07 -.07 -.05 -.02 .02 -.10 Sit. 2 SSI .03 -.03 -.03 .05 .01 -.16 -.01 .10 -.05 .03 SSE .13 -.04 -.05 .02 -.07 -.07 -.25* .16 -.10 .02 Humor .05 .10 .09 .09 .04 .04 .04 .12 -.04 .12 EE .03 -.06 -.03 -.04 .08 -.03 -.05 .14 .07 .07 * SA .13 <.01 .02 .12 -22 .02 -.05 -.11 -.11 -.09 DA .13 .02 >-.01 -.03 -.07 -.07 -.05 -.02 .02 -.10 Sit. 3 SSI -.11 .07 .24* .17* .23* -.22* .14 .10 -.08 .07 SSE .09 .02 .01 .14 -.01 -.07 -.18* .12 -.10 -.12 Humor .10 .04 .05 .01 -.10 .15 -.02 .10 .02 -.06 EE -.04 -.01 .07 .03 .10 -.03 .04 .14 .10 -.07 * * SA .20 -.06 -.07 -.02 -.19 .03 -.03 -.12 -.15 .01 DA .24* -.03 .04 .11 -.13 -.04 -.02 -.04 .05 -.15 Sit. 4 SSI .04 .08 .09 .13 .04 .01 .03 .02 .10 -.01 * SSE .10 .08 .03 .05 -.09 .02 -.18 .09 -.02 -.07 Humor .06 .04 .01 .07 .01 .19* .06 .09 -.06 -.08 EE -.04 .02 .03 .06 .10 .05 .02 .05 .02 -.01 SA .21* -.05 >-.01 -.11 -.17* >-.01 -.06 -.08 -.05 .04 DA .14 -.09 -.04 .02 -.05 .01 -.09 -.01 .04 -.10 Note. *p<.05. SSI=Social Support Seeking-Instrumental, SSE=Social Support Seeking- PT <.01 .05 -.02 .09 -.13 .24* .07 .02 >-.01 .02 >-.01 .24* <.01 -.04 -.03 .07 -.07 .09 -.01 .05 >-.01 -.01 .06 .16* Emotional. EE=Emotional Expression, SA=Surface Acting, DA=Deep Acting, WG sat.= Workgroup satisfaction, HelpI= Workgroup helping intentions, SE= Self Esteem, Impuls.= Impulsivity, Gend=Gender, PT= Position Tenure. Physical strain was rated on a 5-point frequency scale for how often experiences happened in the past 3 months (less than once per month or never- several times a day). P-E fit was measured using a 3-point scale (disagree- 95 Table 13 (cont’d) undecided-agree). Workgroup helping intentions were measured using a 6 point scale (very probably not- definitely not). Self-esteem was measured using a 4-point scale (not at all true of me-true of me to a great extent). Trait impulsivity was measured using a 4 point frequency scale (rarely/never- almost always). Gender was coded 0=women, 1=men; Race was coded 0=NotWhite, 1= White; Level of Employment was coded 0=3/4 time employment, 1=full time (approximately 40 hours per week); Proximity to Colleagues is coded 1= I work virtually outside of the office, 2= I work in an office, but not in close proximity to my colleagues, and 3= I work in an office in close physical proximity to my colleagues; Work experience and tenure in current position were measured in year. Situation 1 was “You discover a colleague has taken credit for your work without permission.” Situation 2 was “A colleague gives you incorrect information about a meeting, causing you to show up an hour late.” Situation 3 was “A reward you were expecting (promotion, raise, bonus, etc. ) was not given to you as promised by the organization.” Situation 4 was “You learn a mistake you made at work resulted in additional work for several other employees.” Situation 5 was “You hear rumors that there will be a layoff in your department.” 96 Table 13 (cont’d) Typical Behavior Sit 5 SSI SSE Humor EE SA DA Phys. Strain P-E Fit WG Sat. HelpI .11 .15 .04 .09 .10 .10 -.04 -.01 .05 -.04 -.07 -.07 .09 -.03 .10 .03 -.07 .04 .02 .10 .04 -.02 >-.01 .02 <.01 .02 .05 -.16 .12 -.07 SE 97 Impuls. Gend Race -.19* -.01 -.06 -.20* .01 .05 -.10 -.11 .10 .02 -.01 -.09 .09 .18* .10 .13 -.11 .04 Work Exp. Phys. PT Prox. .10 -.11 -.05 .06 -.20* .05 .11 -.03 <.01 .08 -.18* -.21* .04 .03 -.07 .07 -.08 .19* Table 14 Descriptive Statistics, Reliabilities, and Intercorrelations for Measures Variable Norms 1.Social Support-Instrumental 2. Social Support- Emotional 3. Use of Humor 4. Emotional Expression 5. Surface Acting 6. Deep Acting 7. Display Rules Coping Behaviors 8.Social Support-Instrumental 9. Social Support- Emotional 10. Use of Humor 11. Emotional Expression 12. Surface Acting 13. Deep Acting Other Variables 14. Physical Strain 15. Workgroup P-E Fit 16. Workgroup Satisfaction 17. Workgroup Helping Intentions 18. Self Esteem 19. Trait Impulsivity 20. Gender 21. Race 22. Work Experience 23. Proximity to Colleagues 24. Tenure in Current Position Items M SD 1. 4 4 3 7 3 3 3 3.82 3.71 3.47 3.27 2.80 2.92 2.86 0.53 (.76) 0.74 .17* (.85) 0.81 .21* -.06 (.89) 0.77 .25* .49* -.03 (.90) 0.77 -.27* -.28* .06 -.69* (.71) 0.82 .12 .22* -.01 .28* -.09 (.83) 0.96 .10 -.13 .07 -.04 .22* -.13 (.82) 20 20 15 21 15 15 3.36 3.06 2.42 3.07 2.64 2.78 0.70 0.97 0.81 0.77 0.83 0.96 .04 .36* -.08 .29* -.21* .11 .19* -.09 .24* -.09 .09 .15 .00 .00 -.04 .31* -.16 .08 .02 .08 .05 .39* -.32* .21* -.31* .34* -.19* .16 -.11 .12 .14 .01 .12 -.14 .21* .11 13 6 17 6 10 30 1 1 1 1 1 1.71 2.36 3.64 4.31 3.34 1.89 0.62 0.78 16.36 2.64 4.77 0.48 -.15 -.02 0.68 -.15 -.05 0.73 .02 -.03 0.73 .02 .02 * 0.43 .18 -.02 0.29 -.27* .04 0.49 -.11 -.21* 0.41 .05 .11 9.22 .10 .01 0.68 .09 .21* 4.01 .02 .05 -.11 .01 .16 .07 .21* -.09 .06 -.01 -.11 -.02 -.12 -.03 .10 .02 -.01 -.02 -.06 .06 -.13 .22* -.26* -.14 .23* -.25* .24* -.04 -.12 .08 -.06 .07 -.11 .03 .01 .14 .00 .15 -.27* .05 .03 * -.20 .15 -.03 -.10 .14 .07 -.05 .14 -.07 .01 -.21* -.04 -.01 .06 -.27* -.10 .12 .16* .03 .01 -.08 .05 -.08 .08 -.07 .04 .03 .21* -.04 .03 .01 .03 .15 2. 98 3. 4. 5. 6. -.09 .10 .15 .08 -.06 .03 .00 .01 .18* .18* .09 7. 8. 9. 10. (.90) .49* (.96) .28* .22* (.94) .37* .57* .25* -.14 -.22* -.06 .32* .25* .14 .06 .07 .07 .07 .01 .16 .06 .14 -.04 .05 -.02 Table 14 (cont’d) Note. *p<.05. Reliability coefficients are in parentheses on the diagonal. Norms and satisfaction with coworkers were rated on a 5point scale of agreement (strongly disagree-strongly agree). Typical behaviors were rated on a 5-point frequency (never-almost always). Physical strain was rated on a 5-point frequency scale for how often experiences happened in the past 3 months (less than once per month or never- several times a day). P-E fit was measured using a 3-point scale (disagree-undecided-agree). Workgroup helping intentions were measured using a 6 point scale (very probably not- definitely not). Self-esteem was measured using a 4-point scale (not at all true of me-true of me to a great extent). Trait impulsivity was measured using a 4 point frequency scale (rarely/neveralmost always). Gender was coded 0=women, 1=men; Race was coded 0=Not-White, 1= White; Level of Employment was coded 0=3/4 time employment, 1=full time (approximately 40 hours per week); Proximity to Colleagues is coded 1= I work virtually outside of the office, 2= I work in an office, but not in close proximity to my colleagues, and 3= I work in an office in close physical proximity to my colleagues; Work experience and tenure in current position were measured in years. 99 Table 14 (cont’d) Variable Coping Behaviors 11. Emotional Expression 12. Surface Acting 13. Deep Acting Other Variables 14. Physical Strain 15. Workgroup P-E Fit 16. Workgroup Satisfaction 17. Workgroup Helping Intentions 18. Self Esteem 19. Trait Impulsivity 20. Gender 21. Race 22. Work Experience 23. Proximity to Colleagues 24. Tenure in Current Position Items M SD 11. 21 15 15 3.07 2.64 2.78 0.77 (.96) 0.83 -.77* (.93) 0.96 .20* .10 (.96) 13 6 17 6 10 30 1 1 1 1 1 1.71 2.36 3.64 4.31 3.34 1.89 0.62 0.78 16.36 1.36 4.77 0.48 0.68 0.73 0.73 0.43 0.29 0.49 0.41 9.22 0.68 4.01 .01 -.02 .02 .01 .10 -.03 -.05 .13 .09 .02 .06 100 12. .18* -.04 -.03 .01 -.21* .04 -.01 -.12 -.18* .04 -.06 13. .19* -.06 -.02 .07 -.12 -.04 -.08 -.01 .04 .17* .19* 14. 15. 16. 17. 18. 19. 20. (.83) -.16 (.92) -.30** .75* (.94) -.08 .22* .37* (.92) -.34* .16* .19* .15 (.80) .32* -.05 -.21* -.14 -.34* (.81) -.22* -.02 .11 .10 .01 .02 n/a .01 .05 .06 -.07 -.05 -.02 .00 -.08 .10 .12 .03 .08 -.14 .06 .02 .01 .05 .23* -.10 -.03 -.08 -.04 .16 .09 .01 .00 -.12 .05 Table 14 (cont’d) Variables 21. Race 22. Work Experience 23. Proximity to Colleagues 24. Tenure in Current Position Items 1 1 1 1 M 0.78 16.36 1.36 4.77 SD 0.41 9.22 0.68 4.01 101 21. n/a -.05 .04 .20* 22. 23. 24. n/a -.06 .40* n/a .03 n/a Table 15 Emotion-Focused Coping Behaviors Regressed on Workgroup Emotion-Focused Coping Norms Criterion Predictor SE B β R2 Social Support Seeking-Instrumental Coping Behavior Social Support Seeking-Instrumental Norm 0.11 0.28 .21** .05 Social Support Seeking-Emotional Coping Behavior Social Support Seeking-Emotional Norm 0.10 0.48 .36** .13 Use of Humor Coping Behavior Use of Humor Norm 0.08 0.24 .24** .06 Emotional Expression Coping Behavior Emotional Expression Norm 0.08 0.40 .39** .15 Surface Acting Behavior Surface Acting Norm 0.08 0.36 .34** .11 Deep Acting Behavior Deep Acting Norm 0.10 0.14 .12 .01 Note. *p<.05, **.01 -.22* >-.01 .01 Step 2 Gender X .17 -.06 EE Norms Note. *p < .05, N= 148 SE B β .12 -.02 -.03 .07 .12 .16 SE B β .12 .08 .15 .02 -.01 -.01 ΔR² .10 .02 .01 <.01 .15 112 Emotional Expression Norms predicting Workgroup Helping Intentions <.01 .01 .11 ΔR² -.02 -.02 .05* <.01 -.05 Emotional Expression Norms predicting Workgroup P-E Fit <.01 .17 -.05 -.05 Table 26 Hypotheses and whether the data supported each Hypothesis Finding Hypothesis 1: Emotion focused coping norms will predict typical emotion-focused coping behaviors. Supported Hypothesis 2: Typical coping through the use of humor and through emotional expression will predict positive distal outcomes. Not Supported Hypothesis 3: Typical coping behaviors will mediate the relationships between their respective norms and distal outcomes. Not Supported Hypothesis 4: The relationships between emotion-focused coping norms and emotion-focused coping behaviors will be moderated by self-esteem, such that low self-esteem will strengthen the relationship. Not Supported Hypothesis 5: The relationships between emotion-focused coping norms and emotion-focused coping behaviors will be moderated by trait impulsivity, such that low trait impulsivity will strengthen relationships. Partially Supported Hypothesis 6: Women will have stronger positive relationships between emotional expression behaviors and positive outcomes than men. Not Supported 113 Table 27 Coping Behaviors Regressed on EFC Norms with Physical Proximity as a Moderator Variables Social Support SeekingInstrumental SE Step 1 Physical Proximity Coping Norm B β ΔR² Social Support SeekingEmotional SE B β .06* 0.08 0.10 .09 0.11 0.29 .22* ΔR² Use of Humor SE B β .13* 0.11 -.01 -.01 Emotional Expression ΔR² SE B β .06* .16* 0.10 -0.07 -.06 0.09 0.01 .01 0.08 0.10 0.47 .36* 0.08 0.40 .39* 0.24 .24* Step 2 <.01 <.01 <.01 Physical Proximity 0.14 -.01 -.01 0.14 -.12 -.07 0.13 -0.08 -.05 0.12 <0.01 <.01 X Coping Norm Note. *p < .05, N= 149. Typical EFC coping behaviors for each regression matched the subdimension of EFC norms (e.g., social support seeking-instrumental norms with typical social support seeking-instrumental behaviors) 114 ΔR² <.01 Table 28 Surface and Deep Acting Behaviors Regressed on Surface Acting, Deep Acting, and Display Rules with Physical Proximity as Moderator Variables Surface Acting Norms predicting Surface Acting Behaviors SE B β ΔR² Step 1 Physical 0.10 -.10 -.08 Proximity Norms 0.08 0.37 .34* SE B β .12* Step 2 Physical Proximity 0.13 0.08 X Norms Deep Acting Norms predicting Deep Acting Behaviors <.01 .05 ΔR² Display Rules predicting Surface Acting Behaviors SE B β 0.04 ΔR² Display Rules predicting Deep Acting Behaviors SE B β 0.05* ΔR² 0.04* 0.12 -.22 -.16 0.10 -.04 -.03 0.12 -.24 -.17* 0.10 0.11 .09 0.07 0.18 .21* 0.08 0.11 .11 0.00 0.14 0.00 <.01 0.01 0.12 0.13 Note. *p < .05, N= 149 115 .09 0.00 -.04 0.14 -.03 Figure 1 Model of the Influence of Workgroup Norms for Emotion Focused Coping on Employee Outcomes Shared workgroup norms for emotionfocused coping Perceived workgroup norms for emotionfocused coping Workgroup norm for coping with negative emotions through emotional expression Trait Impulsivity Self-Esteem Gender Distal Outcomes Emotion-focused Behavioral Outcomes Coping through Emotional Expression Workgroup norms for coping with negative emotions through humor Workgroup norms for coping with negative emotions through social support seeking Workgroup Satisfaction Coping through Humor Physical Strain Symptoms Coping through Social Support Seeking Person-Environment Workgroup Fit Workgroup Helping Intentions Note: The dashed arrow and box show shared norms, which arise from the individual norm perceptions and are measured through aggregation. Due to the nature of the data obtained in the proposed study, this will not be assessed in the current study and is included in this model for conceptual purposes. In this model, emotional suppression is considered to be the lowest end of the continuum of emotional expression. 116 Figure 2 Moderating effect of trait impulsivity on the relationship between norms for emotion-focused Typical Emotion-Focused Support Seeking social support seeking and emotion-focused social support seeking 5 4.5 4 3.5 3 Low Trait Impulsivity 2.5 High Trait Impulsivity 2 1.5 1 Low Emotion-Focused Social Support Seeking Norm High Emotion-Focused Social Support Seeking Norm 117 REFERENCES 118 REFERENCES Affleck, G., & Tennen, H. (1996). Construing benefits from adversity: Adaptational significance and dispositional underpinnings. Journal of Personality, 64, 899-922 Allen, J. A., Pugh, S. D., Grandey, A. A., & Groth, M. (2010). Following display rules in good or bad faith?: Customer orientation as a moderator of the display rule-emotional labor relationship. Human Performance, 23, 101-115. Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labour in service roles: The influence of identity. Academy of Management Review, 18, 88-116. Ashforth, B. E., & Humphrey, R. H. (1995). Emotion in the workplace: A reappraisal. Human Relations, 48, 97-125. Banaji, M. R., & Prentice, D. A. (1994). The self in social contexts. Annual Review of Psychology, 45(1), 297-332. Barratt, E. S. (1959). Anxiety and impulsiveness related to psychomotor efficiency. Perceptual and Motor Skills, 9, 191–198. Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47, 644-675. Barsade, S. G., & Gibson, D. E. (2007). Why does affect matter in organizations? The Academy of Management Perspectives, 21, 36-59. Beal, D. J., Weiss, H. M., Barros, E., & MacDermid, S. M. (2005). An episodic process model of affective influences on performance. Journal of Applied Psychology, 90, 1054-68. Blascovich, J., & Tomaka, J. (1991). Measures of self-esteem. Measures of personality and social psychological attitudes, 1, 115-160. Brotheridge, C. M., & Grandey, A. A. (2002). Emotional labor and burnout: Comparing two perspectives of “people work”. Journal of Vocational Behavior, 60, 17-39. Brotheridge, C. M., & Lee, R. T. (2003). Development and validation of the emotional labour scale. Journal of Occupational and Organizational Psychology, 76, 365-379. Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: A new source of inexpensive, yet high-quality data? Perspectives on Psychological Science, 6, 3-5. Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: a theoretically based approach. Journal of Personality and Social Psychology, 56, 267-283. Cialdini, R. B., & Goldstein, N. J. (2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591-621. 119 Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015-1026. Connor-Smith, J. K., & Flachsbart, C. (2007). Relations between personality and coping: A meta-analysis. Journal of Personality and Social Psychology, 93, 1080-107. Conway, P., & Peetz, J. (2012). When does feeling moral actually make you a better person? Conceptual abstraction moderates whether past moral deeds motivate consistency or compensatory behavior. Personality and Social Psychology Bulletin, 38, 907-919. Danna, K., & Griffin, R. W. (1999). Health and well-being in the workplace: A review and synthesis of the literature. Journal of Management, 25, 357-384. Danoff-Burg, S., Prelow, H. M., & Swenson, R. R. (2004). Hope and life satisfaction in Black college students coping with race-related stress. Journal of Black Psychology, 30, 208228. Dasborough, M. T. (2006). Cognitive asymmetry in employee emotional reactions to leadership behaviors. The Leadership Quarterly, 17, 163-178. DeBono, A., Shmueli, D., & Muraven, M. (2011). Rude and inappropriate: The role of selfcontrol in following social norms. Personality and Social Psychology Bulletin, 37, 136146. Diefendorff, J. M., Erickson, R. J., Grandey, A., & Dahling, J. J. (2011). Emotional display rules as work unit norms: A multilevel analysis of emotional labor among nurses. Journal of occupational health psychology, 16, 170 -186. doi:10.1037/a0021725 Diefendorff, J. M., & Richard, E. M. (2003). Antecedents and consequences of emotional display rule perceptions. Journal of Applied Psychology, 88, 284-294. doi:10.1037/00219010.88.2.284 Diefendorff, J. M., Richard, E. M., & Croyle, M. H. (2006). Are emotional display rules formal job requirements? Examination of employee and supervisor perceptions. Journal of Occupational and Organizational Psychology, 79, 273-298. Diener, E., & Emmons, R. A. (1984). The independence of positive and negative affect. Journal of Personality and Social Psychology, 47, 1105-1117. Ekman, P. (1992). An argument for basic emotions. Cognition & Emotion, 6, 169-200. Fiebig, G. V., & Kramer, M. W. (1998). A framework for the study of emotions in organizational contexts. Management Communication Quarterly, 11, 536-572. Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239. 120 Ganster, D. C., & Schaubroeck, J. (1991). Work stress and employee health. Journal of Management, 17, 235-271. Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85, 348-362. Grosserand, R. H., & Diefendorff, J. M. (2005). Emotional display rules and emotional labor: the moderating role of commitment. Journal of Applied Psychology, 90, 1256. Hunter, E. M., Rubino, C., Perry, S. J., & Penney, L. M. (2013). Expanding the generalizability of surface and deep acting across industries. Poster presented at the annual meeting of the Society for Industrial and Organizational Psychology, Houston, TX. Hui, C., & Lee, C. (2000). Moderating effects of organization-based self-esteem on organizational uncertainty: Employee response relationships. Journal of Management, 26, 215-232. Keltner, D., Ellsworth, P. C., & Edwards, K. (1993). Beyond simple pessimism: Effects of sadness and anger on social perception. Journal of Personality and Social Psychology, 64, 740. Kowalski, R. M. (1996). Complaints and complaining: Functions, antecedents, and consequences. Psychological Bulletin, 119, 179-196. Kozlowski, S. W. J., & Bell, B. S. (2003). Work groups and teams in organizations. In W. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of psychology Industrial and organizational psychology (Vol. 12, pp. 333-375). Kozlowski, S. W. J., Gully, S. M., Nason, E. R., & Smith, E. M. (1999). Developing adaptive teams: A theory of compilation and performance across levels and time. In D. R. Ilgen and E. D. Pulakos (Eds.), The changing nature of work performance: Implications for staffing, personal actions, and development (240-292). San Francisco: Jossey-Bass. Latané, B. (1981). The psychology of social impact. American Psychologist, 36, 343-356. American Psychological Association. Lazarus, R.S. (1966) Psychological stress and the coping process. New York, NY, US: McGraw-Hill. Lazarus, R.S. (1991) Emotion and adaptation. New York: Oxford University Press. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, Springer. Martin, R. A., & Lefcourt, H. M. (1983). Sense of humor as a moderator of the relation between stressors and moods. Journal of Personality and Social Psychology, 45, 1313. MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological methods, 4, 84-99. 121 Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill. Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior, 19, 2-21. Penley, J. A., Tomaka, J., & Wiebe, J. S. (2002). The association of coping to physical and psychological health outcomes: A meta-analytic review. Journal of Behavioral Medicine, 25, 551-603. Pennebaker, J. W., & Chung, C. K. (2011). Expressive writing and its links to mental and physical health. Oxford Handbook of Health Psychology, 417-437. Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. Pratt, T. C., & Cullen, F. T. (2000). The empirical status of Gottfredson and Hirschi's general theory of crime: A meta‐analysis. Criminology, 38, 931-964. Prentice, D. A., & Miller, D. T. (1993). Pluralistic ignorance and alcohol use on campus: Some consequences of misperceiving the social norm. Journal of Personality and Social Psychology, 64(2), 243-56. Rafaeli, A., & Sutton, R. I. (1987). Expression of Emotion as Part of the Work Role. Academy of Management Review, 12(1), 23. Ready, R. E., Clark, L. A., Watson, D., & Westerhouse, K. (2000). Self-and peer-reported personality: Agreement, trait ratability, and the “self-based heuristic”. Journal of Research in Personality, 34(2), 208-224. Romero, E. J., & Cruthirds, K. W. (2006). The use of humor in the workplace. The Academy of Management Perspectives, 20, 58-69. Rosenberg, M. (1965). Society and the adolescent child. Princeton, NJ: Princeton University Press. Roznowski, M. (1989). Examination of the measurement properties of the Job Descriptive Index with experimental items. Journal of Applied Psychology, 74, 805-814. Schaubroeck, J., & Jones, J. R. (2000). Antecedents of workplace emotional labor dimensions and moderators of their effects on physical symptoms. Journal of Organizational Behavior, 21, 163-183. Schneider, B., Salvaggio, A. N., & Subirats, M. (2002). Climate strength: A new direction for climate research. Journal of Applied Psychology, 87, 220-229. Shaver, P., Schwartz, J., Kirson, D., & OʼConnor, C. (1987). Emotion knowledge: further exploration of a prototype approach. Journal of Personality and Social Psychology, 52, 1061-86. 122 Seong, J. Y., & Kristof-Brown, A. L. (2012). Testing multidimensional models of person-group fit. Journal of Managerial Psychology, 27, 536-556. Skinner, E., Edge, K., Altman, J., & Sherwood, H. (2003). Searching for the structure of coping: A review and critique of category systems for classifying ways of coping. Psychological Bulletin, 129, 216-269. Spector, P. E. & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal Conflict at Work Scale, Organizational Constraints Scale, Quantitative Workload Inventory, and Physical Symptoms Inventory. Journal of Occupational Health Psychology, 3, 356-367. Stanford, M. S., Mathias, C. W., Dougherty, D. M., Lake, S. L., Anderson, N. E., & Patton, J. H. (2009). Fifty years of the Barratt Impulsiveness Scale: An update and review. Personality and Individual Differences, 47, 385-395. Stanton, A. L., Kirk, S. B., Cameron, C. L., & Danoff-Burg, S. (2000). Coping through emotional approach: scale construction and validation. Journal of Personality and Social Psychology, 78, 1150. Stanton, A. L., & Low, C. A. (2012). Expressing emotions in stressful contexts benefits, moderators, and mechanisms. Current Directions in Psychological Science, 21, 124-128. Tellegen, A., Watson, D., & Clark, L. (1999). On the dimensional and hierarchical structure of affect. Psychological Science, 10, 297-303. doi:10.1111/1467-9280.00157 Tuckman, B. W. (1965). Developmental sequence in small groups. Psychological Bulletin, 63, 384-399. E Tuckman, B. W., & Jensen, M. A. C. (1977). Stages of small group development revisited. Group and Organization Studies, 2, 419-427. Van Daalen, G., Willemsen, T. M., & Sanders, K. (2006) Reducing work-family conflict through different sources of social support. Journal of Vocational Behavior, 69, 462-476. Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal,41, 108-119. Van Wijk, C. M., & Kolk, A. M. (1997). Sex differences in physical symptoms: The contribution of symptom perception theory. Social Science & Medicine. Volkema, R. J., Farquhar, K., & Bergmann, T. J. (1996). Third-party sensemaking in interpersonal conflicts at work: A theoretical framework. Human Relations, 49, 14371454 Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin, 96, 465-490. 123 Weiss, H. M., & Cropanzano, R. (1996). An affective events approach to job satisfaction. Research In Organizational Behavior, 18, 1-74. 124